# sensitive data

Sensitive data is personal data belonging to special categories that receive enhanced protection, including information about health, sexual life or orientation, racial or ethnic origin, political or religious beliefs, biometric or genetic data, and trade union membership.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## Signal Restricts Screenshots to Block AI Agents From Reading Messages

DevFeed: [Signal Restricts Screenshots to Block AI Agents From Reading Messages](<https://devfeed.tech/articles/signal-restricts-screenshots-to-block-ai-agents-from-reading-messages-146805.md>)

Original publisher: [Read original article](<https://yro.slashdot.org/story/26/11/20/2231247/signal-restricts-screenshots-to-block-ai-agents-from-reading-messages>)

Author: BeauHD

Published: 2026-11-20T23:00:00Z

Content type: news

Language: en

Sources: [Slashdot](<https://devfeed.tech/sources/slashdot.md>)

Topics: [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [AI Bots](<https://devfeed.tech/topics/ai-bots.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [apple](<https://devfeed.tech/tags/apple.md>), [google](<https://devfeed.tech/tags/google.md>), [macos](<https://devfeed.tech/tags/macos.md>), [messaging](<https://devfeed.tech/tags/messaging.md>), [microsoft](<https://devfeed.tech/tags/microsoft.md>), [operating-system](<https://devfeed.tech/tags/operating-system.md>), [privacy](<https://devfeed.tech/tags/privacy.md>), [screenshots](<https://devfeed.tech/tags/screenshots.md>), [signal](<https://devfeed.tech/tags/signal.md>), [windows](<https://devfeed.tech/tags/windows.md>)

### AI overview

Signal is adding default-on restrictions that block screenshots and detected AI assistant screen access to help keep private conversations from being read. Depending on the device, users may see a pop-up allowing them to disable the block. Signal calls the protections incomplete and asks Apple, Google, and Microsoft for clearer ways to prevent agents from accessing sensitive data and more information about how those agents work. Signal previously blocked Microsoft Recall screenshots on Windows, but says it cannot reliably offer comparable protections on macOS.

### Source excerpt

Signal is introducing default-on screenshot restrictions to prevent AI agents from reading private conversations, warning that increasingly powerful assistants integrated into operating systems pose an "existential threat" to its privacy guarantees. "Right now, these agentic AI integrations are changing this environment in a way that's making it increasingly hostile to privacy. This is an existential threat to Signal, and our ability to responsibly uphold our mission," it said in a blog post. Reuters reports: Signal, widely regarded by privacy advocates as a leader in secure messaging, called its new protections "incomplete" and urged operating system providers Apple, Alphabet's Google and Microsoft to give app developers more concrete ways to block the agents from accessing sensitive data. It also called on those companies to provide more information about how their agents function, so developers can assess what data the bots are accessing and adjust their systems accordingly. [...] Users will receive pop-up windows giving them options to disable the blocks when Signal detects screenshot attempts or efforts by an AI assistant to access the screen, depending on the device. Signal similarly blocked Microsoft's Recall AI from taking screenshots on Windows last year. It said it was unable to reliably offer such protections on Apple's macOS. Read more of this story at Slashdot.

## How Bee implemented Zero Operator Access for personal AI on AWS

DevFeed: [How Bee implemented Zero Operator Access for personal AI on AWS](<https://devfeed.tech/articles/how-bee-implemented-zero-operator-access-for-personal-ai-on-aws-146957.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/architecture/how-bee-implemented-zero-operator-access-for-personal-ai-on-aws/>)

Author: Gopinath Jagadesan

Published: 2026-11-20T18:49:05Z

Content type: article

Language: en

Sources: [AWS Architecture Blog](<https://devfeed.tech/sources/aws-architecture-blog.md>)

Topics: [Nitro System](<https://devfeed.tech/topics/nitro-system.md>), [Amazon EC2](<https://devfeed.tech/topics/amazon-ec2.md>), [Amazon VPC](<https://devfeed.tech/topics/amazon-vpc.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [Security, Identity, & Compliance](<https://devfeed.tech/topics/security-identity-compliance.md>), [Amazon Web Services](<https://devfeed.tech/topics/aws.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [amazon-ec2](<https://devfeed.tech/tags/amazon-ec2.md>), [attestation](<https://devfeed.tech/tags/attestation.md>), [aws](<https://devfeed.tech/tags/aws.md>), [bee](<https://devfeed.tech/tags/bee.md>), [customer-solutions](<https://devfeed.tech/tags/customer-solutions.md>), [nitro-system](<https://devfeed.tech/tags/nitro-system.md>), [personal-ai](<https://devfeed.tech/tags/personal-ai.md>)

### AI overview

Bee's Zero Operator Access design gates access to customer data on user-held keys and client verification of server software. At boot, EC2 instance attestation measures the operating system and, through PCR14, container images and model artifacts; expected measurements are independently published to Sigstore/Rekor. The Bee app verifies the attestation and transparency proof in the server's certificate before releasing a short-lived key. Data is encrypted within the attested instance before storage, keys remain in memory, and isolated subnets restrict network paths. Read-only images, disabled interactive access, and AWS organizational controls add safeguards, since attestation alone cannot measure network reach or permission changes. Bee says it plans to move inference to Amazon Bedrock Mantle while retaining user-held keys and attested application servers.

### Source excerpt

Bee, an Amazon company, built a personal AI wearable with no mechanism for operators (Bee engineers, other Amazon employees, or AWS infrastructure operators) to access customer data. Learn how Bee uses Amazon EC2 instance attestation and the AWS Nitro System to achieve Zero Operator Access (ZOA).

## Vendor data breach exposes Chase, Visa, Mastercard customer info

DevFeed: [Vendor data breach exposes Chase, Visa, Mastercard customer info](<https://devfeed.tech/articles/vendor-data-breach-exposes-chase-visa-mastercard-customer-info-146436.md>)

Original publisher: [Read original article](<https://www.bleepingcomputer.com/news/security/vendor-data-breach-exposes-chase-visa-mastercard-customer-info/>)

Author: Bill Toulas

Published: 2026-11-20T15:58:06Z

Content type: news

Language: en

Sources: [BleepingComputer](<https://devfeed.tech/sources/bleepingcomputer.md>)

Topics: [ransomware](<https://devfeed.tech/topics/ransomware.md>), [Security](<https://devfeed.tech/topics/security.md>), [data](<https://devfeed.tech/topics/data.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [Shared Responsibility Model](<https://devfeed.tech/topics/shared-responsibility-model.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [attack](<https://devfeed.tech/tags/attack.md>), [aws](<https://devfeed.tech/tags/aws.md>), [bank](<https://devfeed.tech/tags/bank.md>), [breach](<https://devfeed.tech/tags/breach.md>), [business](<https://devfeed.tech/tags/business.md>), [chase](<https://devfeed.tech/tags/chase.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-data](<https://devfeed.tech/tags/cloud-data.md>), [computer-help](<https://devfeed.tech/tags/computer-help.md>), [computer-security](<https://devfeed.tech/tags/computer-security.md>), [computers](<https://devfeed.tech/tags/computers.md>), [credentials](<https://devfeed.tech/tags/credentials.md>), [customer](<https://devfeed.tech/tags/customer.md>), [customer-data](<https://devfeed.tech/tags/customer-data.md>), [cyberattack](<https://devfeed.tech/tags/cyberattack.md>), [data](<https://devfeed.tech/tags/data.md>), [data-analytics](<https://devfeed.tech/tags/data-analytics.md>), [data-breach](<https://devfeed.tech/tags/data-breach.md>), [exposed](<https://devfeed.tech/tags/exposed.md>), [information](<https://devfeed.tech/tags/information.md>), [infosec](<https://devfeed.tech/tags/infosec.md>), [infosec-computer-security](<https://devfeed.tech/tags/infosec-computer-security.md>), [linux](<https://devfeed.tech/tags/linux.md>), [mac](<https://devfeed.tech/tags/mac.md>), [malware](<https://devfeed.tech/tags/malware.md>), [malware-removal](<https://devfeed.tech/tags/malware-removal.md>), [marketing](<https://devfeed.tech/tags/marketing.md>), [mastercard](<https://devfeed.tech/tags/mastercard.md>), [n0n-ransomware](<https://devfeed.tech/tags/n0n-ransomware.md>), [payment](<https://devfeed.tech/tags/payment.md>), [qubole](<https://devfeed.tech/tags/qubole.md>), [ransomware](<https://devfeed.tech/tags/ransomware.md>), [security](<https://devfeed.tech/tags/security.md>), [spyware](<https://devfeed.tech/tags/spyware.md>), [support](<https://devfeed.tech/tags/support.md>), [tech-support](<https://devfeed.tech/tags/tech-support.md>), [technical-support](<https://devfeed.tech/tags/technical-support.md>), [third-party](<https://devfeed.tech/tags/third-party.md>), [virus](<https://devfeed.tech/tags/virus.md>), [virus-removal](<https://devfeed.tech/tags/virus-removal.md>), [visa](<https://devfeed.tech/tags/visa.md>), [windows](<https://devfeed.tech/tags/windows.md>)

### AI overview

Chase said attackers reached marketing data held by third-party service providers, potentially including customer names and email addresses; the bank reported no service disruption. Visa said a third-party incident in Asia Pacific did not involve its systems or affect payment data. Mastercard reported limited exposure of email addresses for a small subset of cardholders in a specific benefit program, with no effect on its payment network or card use; it said the exposure was contained and it was working with the provider. The excerpt links the incidents to alleged Qubole compromise: attackers reportedly used compromised credentials and preconfigured IAM roles to access customers' AWS environments, while the ransomware group claimed theft of Qubole files and much larger volumes of customer network data. Qubole had not responded by publication.

### Source excerpt

J.P. Morgan Chase warned it suffered a data breach after a cyberattack at a third-party service provider exposed customers' personal information and marketing data. [...]

## 265,000 Oyster photocard customers' data exposed in Transport for London breach

DevFeed: [265,000 Oyster photocard customers' data exposed in Transport for London breach](<https://devfeed.tech/articles/265-000-oyster-photocard-customers-data-exposed-in-transport-for-london-breach-146073.md>)

Original publisher: [Read original article](<https://www.helpnetsecurity.com/2026/11/20/transport-for-london-oyster-photocard-data-breach/>)

Author: Sinisa Markovic

Published: 2026-11-20T10:21:41Z

Content type: news

Language: en

Sources: [Help Net Security](<https://devfeed.tech/sources/help-net-security.md>)

Topics: [Security](<https://devfeed.tech/topics/security.md>), [enterprise data protection](<https://devfeed.tech/topics/enterprise-data-protection.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [account](<https://devfeed.tech/tags/account.md>), [breach](<https://devfeed.tech/tags/breach.md>), [children](<https://devfeed.tech/tags/children.md>), [customers](<https://devfeed.tech/tags/customers.md>), [cyberattack](<https://devfeed.tech/tags/cyberattack.md>), [data](<https://devfeed.tech/tags/data.md>), [data-breach](<https://devfeed.tech/tags/data-breach.md>), [details](<https://devfeed.tech/tags/details.md>), [don-t-miss](<https://devfeed.tech/tags/don-t-miss.md>), [eu](<https://devfeed.tech/tags/eu.md>), [exposed](<https://devfeed.tech/tags/exposed.md>), [hot-stuff](<https://devfeed.tech/tags/hot-stuff.md>), [include](<https://devfeed.tech/tags/include.md>), [london](<https://devfeed.tech/tags/london.md>), [news](<https://devfeed.tech/tags/news.md>), [offline](<https://devfeed.tech/tags/offline.md>), [oyster](<https://devfeed.tech/tags/oyster.md>), [security](<https://devfeed.tech/tags/security.md>), [site](<https://devfeed.tech/tags/site.md>), [system](<https://devfeed.tech/tags/system.md>), [tfl](<https://devfeed.tech/tags/tfl.md>), [transport](<https://devfeed.tech/tags/transport.md>), [transport-for-london](<https://devfeed.tech/tags/transport-for-london.md>), [transportation](<https://devfeed.tech/tags/transportation.md>), [uk](<https://devfeed.tech/tags/uk.md>), [website](<https://devfeed.tech/tags/website.md>), [zip](<https://devfeed.tech/tags/zip.md>)

### AI overview

TfL says a breach of its Oyster photocard website may have exposed account details belonging to about 265,000 holders, including Zip and 60+ cardholders. The potentially compromised information includes email addresses, phone numbers, home addresses, dates of birth, and security questions and answers. TfL's current assessment says passwords, payment details, journey information, children's names and birth dates, photographs, and supporting documents were not compromised; it says other TfL services were unaffected. TfL took the photocard system offline, notified the Information Commissioner's Office, and is investigating with partners. Accounts remain locked; affected users will be asked to change their password and security question and answer when the portal returns.

### Source excerpt

Account details belonging to some 265,000 Oyster photocard holders may have been compromised in a breach of the photocard website, Transport for London (TfL) has warned. They include children with Zip cards and Londoners with the 60+ London Oyster photocard. TfL says it spotted suspicious activity on the site on 16 November and took the photocard system offline to limit access. "While we acted quickly to restrict this, this activity has unfortunately resulted in unauthorised ... More -> The post 265,000 Oyster photocard customers' data exposed in Transport for London breach appeared first on Help Net Security.

## Differentially Private Score Estimation with Applications to Score-Based Generative Models

DevFeed: [Differentially Private Score Estimation with Applications to Score-Based Generative Models](<https://devfeed.tech/articles/differentially-private-score-estimation-with-applications-to-score-based-generative-models-145067.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25536>)

Author: Mengxiao Gao, Elly K. H. Hung, Yuchen Wu, Yi Yu

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [AI Development](<https://devfeed.tech/topics/ai-development.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>)

Tags: [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [differential-privacy](<https://devfeed.tech/tags/differential-privacy.md>), [diffusion](<https://devfeed.tech/tags/diffusion.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [generative-models](<https://devfeed.tech/tags/generative-models.md>), [math-st](<https://devfeed.tech/tags/math-st.md>), [mechanism](<https://devfeed.tech/tags/mechanism.md>), [minimax](<https://devfeed.tech/tags/minimax.md>), [monte-carlo](<https://devfeed.tech/tags/monte-carlo.md>), [score](<https://devfeed.tech/tags/score.md>), [stat-me](<https://devfeed.tech/tags/stat-me.md>), [stat-ml](<https://devfeed.tech/tags/stat-ml.md>), [stat-th](<https://devfeed.tech/tags/stat-th.md>)

### AI overview

The paper presents a differentially private score estimator for d-dimensional sub-Gaussian distributions with locally γ-Hölder-continuous scores. It combines quadratic B-splines with the exponential mechanism and uses projected Langevin Monte Carlo for a polynomial-in-sample-size runtime. Its expected density-weighted squared error is bounded by a polylogarithmic factor times max(n^(-2γ/(d+2γ+2)), (nα)^(-2γ/(d+γ+1))); a nearly matching minimax lower bound characterizes the privacy-accuracy trade-off. For diffusion models, the authors report a privacy guarantee independent of the number of denoising steps and, with suitable discretization, nearly minimax-optimal expected total variation error among private distribution estimators. Experiments on synthetic and real-world datasets are also reported.

### Source excerpt

arXiv:2611.25536v1 Announce Type: cross Abstract: Estimating the score function of a distribution plays a central role in statistics and machine learning, with applications in distribution estimation, sampling, and score-based generative modeling. However, learning scores from sensitive data raises privacy concerns. These concerns are particularly pronounced in score-based generative modeling, where trained models may memorize individual training examples and expose sensitive information through membership inference and data extraction attacks. Although non-private score estimation has been extensively studied, the statistical limits of its private counterpart remain less understood. In this work, we study score estimation under $(\alpha,\beta)$-differential privacy for $d$-dimensional sub-Gaussian distributions with locally $\gamma$-H\"older-continuous scores. We develop a sampling-based private score estimator using quadratic B-splines and the exponential mechanism, and we implement our proposal through projected Langevin Monte Carlo with runtime polynomial in the sample size $n$. The proposed estimator satisfies $(\alpha,\beta)$-differential privacy, and achieves expected density-weighted squared error $\mbox{polylog}(n) \max(n^{-2\gamma/(d+2\gamma+2)},\,(n\alpha)^{-2\gamma/(d+\gamma+1)})$. We also establish a nearly matching private minimax lower bound, thereby characterizing the privacy-accuracy trade-off for score estimation. As an application, we construct a differentially private diffusion model whose privacy guarantee is independent of the number of denoising steps. We show that, with an appropriate discretization scheme, the resulting diffusion model attains nearly minimax optimal expected total variation error among all private distribution estimators. We validate our theoretical findings through experiments on both synthetic and real-world datasets.

## Map sensitive data with Datadog Data Security Posture Management

DevFeed: [Map sensitive data with Datadog Data Security Posture Management](<https://devfeed.tech/articles/map-sensitive-data-with-datadog-data-security-posture-management-147018.md>)

Original publisher: [Read original article](<https://www.datadoghq.com/blog/data-security-posture-management/>)

Author: Loïc Bruneau; Kevin Hu; Alexa Levine; Candace Shamieh

Published: 2026-11-20T00:00:00Z

Content type: article

Language: en

Sources: [Datadog](<https://devfeed.tech/sources/datadog-the-monitor-blog.md>)

Topics: [datadog](<https://devfeed.tech/topics/datadog.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [Amazon S3](<https://devfeed.tech/topics/amazon-s3.md>), [Security](<https://devfeed.tech/topics/security.md>), [Data Management](<https://devfeed.tech/topics/data-management.md>), [rdbms](<https://devfeed.tech/topics/rdbms.md>)

Tags: [amazon-s3](<https://devfeed.tech/tags/amazon-s3.md>), [data-observability](<https://devfeed.tech/tags/data-observability.md>), [data-security](<https://devfeed.tech/tags/data-security.md>), [data-security-posture-management](<https://devfeed.tech/tags/data-security-posture-management.md>), [database-monitoring](<https://devfeed.tech/tags/database-monitoring.md>), [datadog](<https://devfeed.tech/tags/datadog.md>), [dspm](<https://devfeed.tech/tags/dspm.md>), [financial-data](<https://devfeed.tech/tags/financial-data.md>), [map](<https://devfeed.tech/tags/map.md>), [pii](<https://devfeed.tech/tags/pii.md>), [postgresql](<https://devfeed.tech/tags/postgresql.md>), [preview](<https://devfeed.tech/tags/preview.md>), [secrets](<https://devfeed.tech/tags/secrets.md>), [security](<https://devfeed.tech/tags/security.md>), [sensitive-data](<https://devfeed.tech/tags/sensitive-data.md>), [snowflake](<https://devfeed.tech/tags/snowflake.md>)

### AI overview

Datadog DSPM, available in Preview, locates PII, financial data, and secrets at rest in Snowflake, PostgreSQL, and Amazon S3, showing findings in the Data Security explorer and Datadog pages engineers already use. It reuses Sensitive Data Scanner rules and reports the matched rule, location, and count without retaining or displaying sensitive values. Scanning differs by store: Snowflake sends sampled rows to a Datadog-hosted crawler for in-memory scanning; PostgreSQL scans sampled rows in the existing Datadog Agent; and Amazon S3 uses Agentless Scanning to scan sampled files in the customer's AWS account, with bucket restrictions available. PostgreSQL access may require additional table grants, which DSPM identifies; Fleet Automation can configure managed Agents automatically. Supported S3 formats include CSV, JSON, and Parquet. Planned links to lineage, query activity, and AI agent traces are future work, not current Preview capabilities.

### Source excerpt

Datadog DSPM finds PII, financial data, and secrets at rest in Snowflake, PostgreSQL, and Amazon S3, down to the exact column or file. Now in Preview.

## Hackers steal data of 1.6 million SBC Medical Group customers

DevFeed: [Hackers steal data of 1.6 million SBC Medical Group customers](<https://devfeed.tech/articles/hackers-steal-data-of-1-6-million-sbc-medical-group-customers-144567.md>)

Original publisher: [Read original article](<https://www.bleepingcomputer.com/news/security/hackers-steal-data-of-16-million-sbc-medical-group-customers/>)

Author: Bill Toulas

Published: 2026-11-19T22:25:42Z

Content type: news

Language: en

Sources: [BleepingComputer](<https://devfeed.tech/sources/bleepingcomputer.md>)

Topics: [Cybersecurity](<https://devfeed.tech/topics/cybersecurity.md>), [data loss prevention](<https://devfeed.tech/topics/data-loss-prevention.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [data](<https://devfeed.tech/topics/data.md>)

Tags: [administrator](<https://devfeed.tech/tags/administrator.md>), [all](<https://devfeed.tech/tags/all.md>), [bank](<https://devfeed.tech/tags/bank.md>), [basic](<https://devfeed.tech/tags/basic.md>), [booking](<https://devfeed.tech/tags/booking.md>), [breach](<https://devfeed.tech/tags/breach.md>), [computer-help](<https://devfeed.tech/tags/computer-help.md>), [computer-security](<https://devfeed.tech/tags/computer-security.md>), [computers](<https://devfeed.tech/tags/computers.md>), [customer-data](<https://devfeed.tech/tags/customer-data.md>), [customers](<https://devfeed.tech/tags/customers.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [data-breach](<https://devfeed.tech/tags/data-breach.md>), [database](<https://devfeed.tech/tags/database.md>), [employee](<https://devfeed.tech/tags/employee.md>), [employees](<https://devfeed.tech/tags/employees.md>), [exposed](<https://devfeed.tech/tags/exposed.md>), [healthcare](<https://devfeed.tech/tags/healthcare.md>), [incidents](<https://devfeed.tech/tags/incidents.md>), [infosec](<https://devfeed.tech/tags/infosec.md>), [infosec-computer-security](<https://devfeed.tech/tags/infosec-computer-security.md>), [investigation](<https://devfeed.tech/tags/investigation.md>), [japan](<https://devfeed.tech/tags/japan.md>), [linux](<https://devfeed.tech/tags/linux.md>), [mac](<https://devfeed.tech/tags/mac.md>), [malware](<https://devfeed.tech/tags/malware.md>), [malware-removal](<https://devfeed.tech/tags/malware-removal.md>), [medical](<https://devfeed.tech/tags/medical.md>), [patient-data](<https://devfeed.tech/tags/patient-data.md>), [personal-data](<https://devfeed.tech/tags/personal-data.md>), [sbc-medical-group](<https://devfeed.tech/tags/sbc-medical-group.md>), [sec](<https://devfeed.tech/tags/sec.md>), [securities-and-exchange-commission](<https://devfeed.tech/tags/securities-and-exchange-commission.md>), [security](<https://devfeed.tech/tags/security.md>), [spyware](<https://devfeed.tech/tags/spyware.md>), [support](<https://devfeed.tech/tags/support.md>), [tech-support](<https://devfeed.tech/tags/tech-support.md>), [technical-support](<https://devfeed.tech/tags/technical-support.md>), [virus](<https://devfeed.tech/tags/virus.md>), [virus-removal](<https://devfeed.tech/tags/virus-removal.md>), [windows](<https://devfeed.tech/tags/windows.md>)

### AI overview

SBC Medical Group reported two cybersecurity incidents affecting its Japan-based businesses. A breach of databases in a website system exposed personal information associated with 1,624,000 people, though not every listed data type applied to everyone; the compromised data included names, contact details, dates of birth, IDs, passwords and survey responses. The breach also involved reservation requests, website inquiries and about 20,000 employee account records, including bank and tax information for 459 employees. Separately, unauthorized access to administrator accounts on Tribeau, a booking platform used by 56 clinics, exposed reservation information for 12,641 people and treatment details for 6,804. SBC said it had not confirmed access to electronic medical records or data from its mySBC reservation site, and had seen no evidence of misuse as of November 18. The investigation continued, and financial impact had not been estimated.

### Source excerpt

SBC Medical Group has warned in a filing with the U.S. Securities and Exchange Commission (SEC) that two cybersecurity incidents exposed sensitive customer and employee information, including personal data belonging to 1.6 million people. [...]

## Meta's AI agent Muse is mixing up conversations and exposing private data, report says

DevFeed: [Meta's AI agent Muse is mixing up conversations and exposing private data, report says](<https://devfeed.tech/articles/meta-s-ai-agent-muse-is-mixing-up-conversations-and-exposing-private-data-report-says-143920.md>)

Original publisher: [Read original article](<https://techcrunch.com/2026/11/19/metas-ai-agent-muse-is-mixing-up-conversations-and-exposing-private-data-report-says/>)

Author: Aisha Malik

Published: 2026-11-19T17:38:40Z

Content type: news

Language: en

Sources: [TechCrunch | AI News & Artificial Intelligence](<https://devfeed.tech/sources/techcrunch-ai-news-artificial-intelligence.md>)

Topics: [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [Conversational AI](<https://devfeed.tech/topics/conversational-ai.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>)

Tags: [account](<https://devfeed.tech/tags/account.md>), [advice](<https://devfeed.tech/tags/advice.md>), [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [app](<https://devfeed.tech/tags/app.md>), [apps](<https://devfeed.tech/tags/apps.md>), [bank](<https://devfeed.tech/tags/bank.md>), [career](<https://devfeed.tech/tags/career.md>), [career-advice](<https://devfeed.tech/tags/career-advice.md>), [conversations](<https://devfeed.tech/tags/conversations.md>), [data](<https://devfeed.tech/tags/data.md>), [employees](<https://devfeed.tech/tags/employees.md>), [in-brief](<https://devfeed.tech/tags/in-brief.md>), [messages](<https://devfeed.tech/tags/messages.md>), [meta](<https://devfeed.tech/tags/meta.md>), [mixing](<https://devfeed.tech/tags/mixing.md>), [muse](<https://devfeed.tech/tags/muse.md>), [orders](<https://devfeed.tech/tags/orders.md>), [private](<https://devfeed.tech/tags/private.md>), [product](<https://devfeed.tech/tags/product.md>)

### AI overview

Meta employees reportedly raised concerns that its AI agent Muse mixed up conversations across accounts and exposed private information. Examples in internal posts included unrequested childcare and career messages, product orders, bank balances, and a reminder about a Signal message that one employee said she had not authorized Muse to access. The account is based on a Reuters report of internal posts; Meta had not immediately responded to TechCrunch's request for comment. The excerpt also says Muse launched in September and had more than 10 million downloads.

### Source excerpt

In some of these instances, employees said Muse surfaced messages about childcare requests they never made, career advice they never asked for, product orders they never placed, and bank account balances that didn't belong to them.

## A Forensic Analysis of Metadata and Information Leakage in Swedish Public Documents

DevFeed: [A Forensic Analysis of Metadata and Information Leakage in Swedish Public Documents](<https://devfeed.tech/articles/a-forensic-analysis-of-metadata-and-information-leakage-in-swedish-public-documents-141472.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.21922>)

Author: Noor Abo Hasan, Kajsa Linderoth, Monia L\\"onn

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [pdf](<https://devfeed.tech/topics/pdf.md>), [Cryptography](<https://devfeed.tech/topics/cryptography.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>)

Tags: [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [exiftool](<https://devfeed.tech/tags/exiftool.md>), [forensic](<https://devfeed.tech/tags/forensic.md>), [information](<https://devfeed.tech/tags/information.md>), [metadata](<https://devfeed.tech/tags/metadata.md>), [pdf](<https://devfeed.tech/tags/pdf.md>), [public](<https://devfeed.tech/tags/public.md>)

### AI overview

An analysis of 2,876 publicly available PDFs from 30 Swedish public-sector authorities found significant metadata exposure (R2 or R3) in 48.4% of documents, with rates varying across institutional tiers. Exposure declined significantly between the baseline and post-enactment periods for the Swedish Cybersäkerhetslagen, but 44.9% of post-enactment documents still leaked sensitive metadata. Incremental update layers appeared in 41.6% of documents; forensic analysis recovered verifiable text alterations, including deleted personal data and internal identifiers, in 228 cases. The abstract attributes the persistence of leakage to regulatory lag and reliance on default, unsanitized publication workflows.

### Source excerpt

arXiv:2611.21922v1 Announce Type: new Abstract: This study investigates metadata and structural content leakage in publicly available PDF documents from 30 Swedish public-sector authorities. Using an automated collection pipeline and ExifTool-based extraction, 2,876 documents were analyzed and classified using an R0-R3 risk taxonomy. Results reveal that 48.4% of documents exhibit significant metadata exposure (R2 or R3), with significant variation across institutional tiers (p < 0.001). Longitudinal analysis across three legislative periods reveals a statistically significant reduction in metadata exposure between the baseline and post-enactment periods, yet 44.9% of documents published after the enactment of the Swedish Cybers\"akerhetslagen continue to leak sensitive metadata. Furthermore, 41.6% of documents contain incremental update layers; forensic analysis recovered verifiable text alterations, including deleted personal data and internal identifiers, from 228 of these cases. These findings highlight persistent regulatory lag and a systemic reliance on default, unsanitized document publication workflows within the Swedish public sector.

## Flexibility Has a Price: Quantifying Modality Policies in Maliciously Secure Multi-Biometric Authentication

DevFeed: [Flexibility Has a Price: Quantifying Modality Policies in Maliciously Secure Multi-Biometric Authentication](<https://devfeed.tech/articles/flexibility-has-a-price-quantifying-modality-policies-in-maliciously-secure-multi-biometric-authentication-141495.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22714>)

Author: Fuyi Wang, Wang Zhang, Neo Xu, Chao Chen, Kok-Leong Ong, Jianying Zhou

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [Cryptography](<https://devfeed.tech/topics/cryptography.md>), [Access Control](<https://devfeed.tech/topics/access-control.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>)

Tags: [authentication](<https://devfeed.tech/tags/authentication.md>), [biometric-authentication](<https://devfeed.tech/tags/biometric-authentication.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [flexibility](<https://devfeed.tech/tags/flexibility.md>), [policies](<https://devfeed.tech/tags/policies.md>), [privacy](<https://devfeed.tech/tags/privacy.md>), [security](<https://devfeed.tech/tags/security.md>)

### AI overview

Accepting more biometric modality subsets creates more authentication entry points, so the paper treats flexibility as a security policy that must be selected against a target risk and attempt budget. It proposes a policy-selection framework and SUPERMBA, a two-server protocol that hides biometric templates and the subset used, enforces policy verifiably with one potentially malicious server, and performs the same computation across policies and subsets. On MULB+, allowing all subsets raised impostor risk from 1.21% for the weakest entry to 4.42% for their union; on VidTIMIT, the increase was smaller and not statistically resolved with the available identities. Authentication took 6.42 ms and 6.03 ms on those datasets, respectively.

### Source excerpt

arXiv:2611.22714v1 Announce Type: new Abstract: Multi-biometric authentication improves robustness and accessibility by combining information from multiple biometric modalities. Existing research has primarily focused on secure multimodal authentication, feature fusion, and recognition performance, while paying comparatively little attention to the security implications of modality-flexible authentication policies. At the same time, most systems provide limited cryptographic protection for biometric data during authentication. This paper addresses both challenges. We first study modality flexibility as a security decision rather than merely a usability feature. Each accepted modality subset creates an authentication entry point, and the resulting risk depends on the joint behavior of all accepted subsets and the attempt budget. We introduce a framework for quantifying this security cost and develop a policy-selection procedure that identifies policies satisfying a target risk level while supporting users who can present only a subset of modalities. We then present SUPERMBA, a two-server privacy-preserving authentication system that enforces the selected policy without revealing biometric templates or the modality subset used in an authentication attempt. The protocol provides verifiable policy enforcement even when one server is malicious and executes the same computation irrespective of the chosen policy or modality subset. Experiments on the three-modality MULB+ dataset show that allowing all modality subsets increases impostor risk from 1.21% for the weakest entry to 4.42% for the union of entries. On VidTIMIT, the increase is substantially smaller and is not statistically resolved with the available identities. SUPERMBA completes an authentication attempt in only 6.42 ms on MULB+ and 6.03 ms on VidTIMIT.

## Protecting Sensitive Data Across Every Connector | Blog | Fivetran

DevFeed: [Protecting Sensitive Data Across Every Connector | Blog | Fivetran](<https://devfeed.tech/articles/protecting-sensitive-data-across-every-connector-blog-fivetran-140251.md>)

Original publisher: [Read original article](<https://wf.fivetran.com/blog/protecting-sensitive-data-across-every-connector>)

Author: Amy Peterson

Published: 2026-11-18T03:45:08Z

Content type: article

Language: en

Sources: [Fivetran Blog](<https://devfeed.tech/sources/fivetran-blog.md>)

Topics: [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [data loss prevention](<https://devfeed.tech/topics/data-loss-prevention.md>), [Security](<https://devfeed.tech/topics/security.md>)

Tags: [accounting](<https://devfeed.tech/tags/accounting.md>), [apps](<https://devfeed.tech/tags/apps.md>), [cloud-security](<https://devfeed.tech/tags/cloud-security.md>), [column](<https://devfeed.tech/tags/column.md>), [compliance](<https://devfeed.tech/tags/compliance.md>), [connectors](<https://devfeed.tech/tags/connectors.md>), [data](<https://devfeed.tech/tags/data.md>), [data-warehouse](<https://devfeed.tech/tags/data-warehouse.md>), [events](<https://devfeed.tech/tags/events.md>), [files](<https://devfeed.tech/tags/files.md>), [fivetran](<https://devfeed.tech/tags/fivetran.md>), [gdpr](<https://devfeed.tech/tags/gdpr.md>), [masking](<https://devfeed.tech/tags/masking.md>), [pii](<https://devfeed.tech/tags/pii.md>), [saas](<https://devfeed.tech/tags/saas.md>), [sensitive-data](<https://devfeed.tech/tags/sensitive-data.md>)

### AI overview

Fivetran makes column masking available across all connectors, including SaaS applications, files, events and functions, as well as database connectors. In the connector's Schema Tab, users can block a field so it is not loaded into the destination, or hash its values before loading. Blocking makes the data unavailable for querying; hashing anonymizes it and can support analysis across data sources. The text says the feature was rolling out gradually, with earlier access available by contacting Fivetran.

### Source excerpt

Column masking for PII and other sensitive data is now universally available -- including for SaaS apps, files, events and more.

## Cyera Agent Guardian + Microsoft Agent 365: Show What Agents Actually Do With Sensitive Data

DevFeed: [Cyera Agent Guardian + Microsoft Agent 365: Show What Agents Actually Do With Sensitive Data](<https://devfeed.tech/articles/cyera-agent-guardian-microsoft-agent-365-show-what-agents-actually-do-with-sensitive-data-140321.md>)

Original publisher: [Read original article](<https://www.cyera.com/blog/cyera-agent-guardian-microsoft-agent-365>)

Author: Devin Hamilton

Published: 2026-11-17T22:45:51Z

Content type: release

Language: en

Sources: [Cyera Blogs](<https://devfeed.tech/sources/cyera-blogs.md>)

Topics: [data loss prevention](<https://devfeed.tech/topics/data-loss-prevention.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [alert triage](<https://devfeed.tech/topics/alert-triage.md>), [Microsoft](<https://devfeed.tech/topics/microsoft.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-365](<https://devfeed.tech/tags/agent-365.md>), [agent-context](<https://devfeed.tech/tags/agent-context.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-observability](<https://devfeed.tech/tags/ai-observability.md>), [alert](<https://devfeed.tech/tags/alert.md>), [alerts](<https://devfeed.tech/tags/alerts.md>), [analysts](<https://devfeed.tech/tags/analysts.md>), [attribution](<https://devfeed.tech/tags/attribution.md>), [benefits](<https://devfeed.tech/tags/benefits.md>), [cyera](<https://devfeed.tech/tags/cyera.md>), [data](<https://devfeed.tech/tags/data.md>), [microsoft](<https://devfeed.tech/tags/microsoft.md>), [microsoft-agent-365](<https://devfeed.tech/tags/microsoft-agent-365.md>), [microsoft-sentinel](<https://devfeed.tech/tags/microsoft-sentinel.md>), [risk](<https://devfeed.tech/tags/risk.md>), [runtime](<https://devfeed.tech/tags/runtime.md>), [sensitive-data](<https://devfeed.tech/tags/sensitive-data.md>)

### AI overview

Cyera Agent Guardian contributes data-aware runtime findings through Microsoft Sentinel to enrich Microsoft Agent 365's dynamic risk assessment for supported agents. It inspects tool calls, identifies sensitive data, enforces policy inline, and can quarantine a compromised agent. Findings describe the action, data sensitivity and severity, agent and human attribution, and violated policy with Cyera's assessment; alerts retain Cyera as their source and are correlated with the agent. Microsoft combines them with its own runtime alerts to calculate per-agent risk, surfaced in the Microsoft 365 admin center, Defender, Entra, and Purview. The excerpt says analysts can investigate supported Cyera alerts in Defender XDR through Sentinel workflows. The integration is in limited private preview for Cyera customers; access is requested through a Cyera account team.

### Source excerpt

Cyera Agent Guardian sends data-aware runtime findings to Microsoft Sentinel to enrich Agent 365 risk, showing what AI agents actually do with sensitive data.

## Protect sensitive data in your traces: new controls in LangSmith

DevFeed: [Protect sensitive data in your traces: new controls in LangSmith](<https://devfeed.tech/articles/protect-sensitive-data-in-your-traces-new-controls-in-langsmith-139469.md>)

Original publisher: [Read original article](<https://www.langchain.com/blog/protect-sensitive-data-in-your-traces-new-controls-in-langsmith>)

Author: LangChain Accounts

Published: 2026-11-17T21:31:36Z

Content type: release

Language: en

Sources: [LangChain Blog](<https://devfeed.tech/sources/langchain-blog.md>)

Topics: [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [telemetry](<https://devfeed.tech/topics/telemetry.md>), [Data Management](<https://devfeed.tech/topics/data-management.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [about](<https://devfeed.tech/tags/about.md>), [access-control](<https://devfeed.tech/tags/access-control.md>), [access-controls](<https://devfeed.tech/tags/access-controls.md>), [added](<https://devfeed.tech/tags/added.md>), [admin](<https://devfeed.tech/tags/admin.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [agents-in-production](<https://devfeed.tech/tags/agents-in-production.md>), [all](<https://devfeed.tech/tags/all.md>), [controls](<https://devfeed.tech/tags/controls.md>), [data](<https://devfeed.tech/tags/data.md>), [help](<https://devfeed.tech/tags/help.md>), [langsmith](<https://devfeed.tech/tags/langsmith.md>), [manage](<https://devfeed.tech/tags/manage.md>), [more](<https://devfeed.tech/tags/more.md>), [sensitive-data](<https://devfeed.tech/tags/sensitive-data.md>), [teams](<https://devfeed.tech/tags/teams.md>), [trace](<https://devfeed.tech/tags/trace.md>), [traces](<https://devfeed.tech/tags/traces.md>)

### AI overview

LangSmith Enterprise adds three controls for managing access to complete production traces: admins can hide trace inputs and outputs while leaving run trees, performance statistics, and metadata visible; assign workspace-specific permissions to user groups, layered on existing RBAC and ABAC; and define trace-level ABAC policies from root run metadata. Layered together, these let teams grant, for example, support staff access to relevant customers' inputs and outputs while engineers can inspect run and performance data across traces with sensitive content hidden. Policies can be adjusted as team responsibilities change. The text says full-fidelity traces can be stored without redacted or duplicate copies; the controls determine what each user can view. These features are available on LangSmith Enterprise plans.

### Source excerpt

New controls are now in LangSmith to help teams manage access to sensitive trace data. Learn more.

## Italy fines Meta EUR 805 million over a Facebook feature that scanned Android camera rolls

DevFeed: [Italy fines Meta EUR 805 million over a Facebook feature that scanned Android camera rolls](<https://devfeed.tech/articles/italy-fines-meta-805-million-over-a-facebook-feature-that-scanned-android-camera-rolls-138042.md>)

Original publisher: [Read original article](<https://blog.adafruit.com/2026/11/17/italy-fines-meta-e805-million-over-a-facebook-feature-that-scanned-android-camera-rolls/>)

Author: Anne Barela

Published: 2026-11-17T17:09:42Z

Content type: news

Language: en

Sources: [Adafruit Industries](<https://devfeed.tech/sources/adafruit-industries-makers-hackers-artists-designers-and-engineers.md>)

Topics: [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [Android](<https://devfeed.tech/topics/android.md>), [Image](<https://devfeed.tech/topics/image.md>)

Tags: [2023](<https://devfeed.tech/tags/2023.md>), [2025](<https://devfeed.tech/tags/2025.md>), [android](<https://devfeed.tech/tags/android.md>), [android-phones](<https://devfeed.tech/tags/android-phones.md>), [app](<https://devfeed.tech/tags/app.md>), [article](<https://devfeed.tech/tags/article.md>), [camera](<https://devfeed.tech/tags/camera.md>), [code](<https://devfeed.tech/tags/code.md>), [computers](<https://devfeed.tech/tags/computers.md>), [consumer](<https://devfeed.tech/tags/consumer.md>), [data](<https://devfeed.tech/tags/data.md>), [data-protection](<https://devfeed.tech/tags/data-protection.md>), [facebook](<https://devfeed.tech/tags/facebook.md>), [feature](<https://devfeed.tech/tags/feature.md>), [fines](<https://devfeed.tech/tags/fines.md>), [images](<https://devfeed.tech/tags/images.md>), [italy](<https://devfeed.tech/tags/italy.md>), [meta](<https://devfeed.tech/tags/meta.md>), [photos](<https://devfeed.tech/tags/photos.md>), [privacy](<https://devfeed.tech/tags/privacy.md>), [security](<https://devfeed.tech/tags/security.md>), [videos](<https://devfeed.tech/tags/videos.md>)

### AI overview

Italy's Garante Privacy fined Meta EUR 805 million over Facebook's Android camera-roll suggestions feature, which accessed phone photos and videos to propose posts. The regulator found that access relied on a single app-wide permission and unclear notice, and that the feature was enabled by default, violating Article 122 of Italy's Data Protection Code. About 20 million Italian Android users were affected. The GPDP said Meta's lack of cooperation raised the fine, about 0.47% of its global turnover. Meta has 60 days to disable the feature and request fresh Android user permission. Meta says it will appeal and calls the penalty disproportionate.

### Source excerpt

Alfonso Maruccia writes at TechSpot that Italy's data protection authority, the Garante Privacy (GPDP), has fined Meta EUR 805 million, about $887 million, over a Facebook feature that read photos and videos from Android phones. The feature, "Get camera roll suggestions when you're browsing Facebook," has been around since 2023. It pulls images straight from the [...]

## Access Dental Management Data Breach Impacts Over 340,000

DevFeed: [Access Dental Management Data Breach Impacts Over 340,000](<https://devfeed.tech/articles/access-dental-management-data-breach-impacts-over-340-000-137795.md>)

Original publisher: [Read original article](<https://www.securityweek.com/access-dental-management-data-breach-impacts-over-340000/>)

Author: Ionut Arghire

Published: 2026-11-17T12:59:04Z

Content type: news

Language: en

Sources: [SecurityWeek](<https://devfeed.tech/sources/securityweek.md>)

Topics: [Security](<https://devfeed.tech/topics/security.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [network security](<https://devfeed.tech/topics/network-security.md>)

Tags: [access-dental-management](<https://devfeed.tech/tags/access-dental-management.md>), [breach](<https://devfeed.tech/tags/breach.md>), [cyberattack](<https://devfeed.tech/tags/cyberattack.md>), [data-breach](<https://devfeed.tech/tags/data-breach.md>), [data-breaches](<https://devfeed.tech/tags/data-breaches.md>), [data-privacy](<https://devfeed.tech/tags/data-privacy.md>), [hack](<https://devfeed.tech/tags/hack.md>), [hackers](<https://devfeed.tech/tags/hackers.md>), [health](<https://devfeed.tech/tags/health.md>), [healthcare](<https://devfeed.tech/tags/healthcare.md>), [identity](<https://devfeed.tech/tags/identity.md>), [identity-theft](<https://devfeed.tech/tags/identity-theft.md>), [incident](<https://devfeed.tech/tags/incident.md>), [insurance](<https://devfeed.tech/tags/insurance.md>), [medical](<https://devfeed.tech/tags/medical.md>), [more](<https://devfeed.tech/tags/more.md>), [network](<https://devfeed.tech/tags/network.md>), [notification](<https://devfeed.tech/tags/notification.md>), [number](<https://devfeed.tech/tags/number.md>), [personal](<https://devfeed.tech/tags/personal.md>), [personal-information](<https://devfeed.tech/tags/personal-information.md>), [privacy](<https://devfeed.tech/tags/privacy.md>)

### AI overview

Access Dental Management reported that hackers accessed its network from May 1 to May 4 and viewed or copied data. Its review identified names, addresses, birth dates, Social Security numbers, medical information, and health insurance information. ADM reported 341,049 affected people to Texas and Vermont, though the total could reportedly exceed 455,000. It is notifying potentially affected individuals and advises them to watch account statements, insurance explanations of benefits, and credit reports for suspicious activity or errors. ADM says it is reviewing and enhancing privacy policies and procedures. The WorldLeaks ransomware group claimed responsibility, but ADM did not share further attack details.

### Source excerpt

Hackers breached Access Dental's network and stole personal, health, and insurance information. The post Access Dental Management Data Breach Impacts Over 340,000 appeared first on SecurityWeek.

## Quantifying Membership Inference Risks in Genomic Data Sharing: The Role of Attacker's Auxiliary Information

DevFeed: [Quantifying Membership Inference Risks in Genomic Data Sharing: The Role of Attacker's Auxiliary Information](<https://devfeed.tech/articles/quantifying-membership-inference-risks-in-genomic-data-sharing-the-role-of-attacker-s-auxiliary-information-138801.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18909>)

Author: Maryam Ghasemian, Lynette Hammond Gerido, Erman Ayday

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Differential Privacy](<https://devfeed.tech/topics/differential-privacy.md>), [Security & Privacy](<https://devfeed.tech/topics/security-privacy.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [Data analysis](<https://devfeed.tech/topics/data-analysis.md>)

Tags: [attack](<https://devfeed.tech/tags/attack.md>), [auditing](<https://devfeed.tech/tags/auditing.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [differential-privacy](<https://devfeed.tech/tags/differential-privacy.md>), [false-positive-rate](<https://devfeed.tech/tags/false-positive-rate.md>), [framework](<https://devfeed.tech/tags/framework.md>), [genomic-data](<https://devfeed.tech/tags/genomic-data.md>), [local](<https://devfeed.tech/tags/local.md>), [mechanism](<https://devfeed.tech/tags/mechanism.md>), [power](<https://devfeed.tech/tags/power.md>), [pre-release](<https://devfeed.tech/tags/pre-release.md>), [precision](<https://devfeed.tech/tags/precision.md>), [principal-component-analysis](<https://devfeed.tech/tags/principal-component-analysis.md>), [privacy](<https://devfeed.tech/tags/privacy.md>), [project](<https://devfeed.tech/tags/project.md>), [public](<https://devfeed.tech/tags/public.md>), [recall](<https://devfeed.tech/tags/recall.md>), [research](<https://devfeed.tech/tags/research.md>), [sharing](<https://devfeed.tech/tags/sharing.md>)

### AI overview

The proposed framework audits genomic datasets for membership inference risk before publication. It maps data into a shared PCA space, applies epsilon-local differential privacy with Laplace noise using a derived sensitivity for PCA scores, and measures a distance-threshold attack at a 5% false-positive rate. It also reports precision, recall, and an Unbiased Reconstruction Robustness membership-neutrality diagnostic. Experiments on openSNP and 1000 Genomes Project data across dimensionalities and privacy budgets identify separation between the research dataset and public auxiliary data in PCA space as a stronger determinant of attack success than perturbation level alone. A GWAS-style likelihood ratio test provides a comparison baseline.

### Source excerpt

arXiv:2611.18909v1 Announce Type: cross Abstract: Genomic data sharing is essential for biomedical research but introduces significant privacy risks, including membership inference attacks, in which an adversary determines whether a specific individual's data was used in a study. Existing privacy protections are typically applied after a release format has already been chosen, leaving researchers without a principled method to assess dataset-specific risk beforehand. We propose a server-assisted, principal component analysis-based pre-release auditing framework that maps genomic data into a shared low-dimensional space, applies epsilon-local differential privacy via the Laplace mechanism with a formally derived sensitivity for principal component analysis scores, and estimates membership inference attack power at a controlled 5% false-positive rate using a distance-threshold attack. A membership-neutrality diagnostic based on Unbiased Reconstruction Robustness is reported alongside attack power, precision, and recall. Experiments using the openSNP and 1000 Genomes Project datasets across multiple dimensionalities and privacy budgets show that separation between the researcher's dataset and public auxiliary data in principal component analysis space is a dominant determinant of attack success, exceeding the effect of perturbation level alone. Comparison against a genome-wide association study-style likelihood ratio test baseline further contextualizes the measured risk. This framework enables researchers, data repositories, and institutional review bodies to quantify dataset-specific membership inference risk before genomic data publication, supporting more informed and privacy-aware biomedical data-sharing decisions.

## Evidence-Conditioned Completion and Synthesis of Electronic Health Records

DevFeed: [Evidence-Conditioned Completion and Synthesis of Electronic Health Records](<https://devfeed.tech/articles/evidence-conditioned-completion-and-synthesis-of-electronic-health-records-137460.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19546>)

Author: Sicheng Shu, Parker Knight, Huitao Li, Yuan Luo, Yifan Peng, Shawn N. Murphy, Rui Yang, Chuan Hong

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [AI Foundation Models](<https://devfeed.tech/topics/ai-foundation-models.md>), [Representation learning](<https://devfeed.tech/topics/representation-learning.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [Differential Privacy](<https://devfeed.tech/topics/differential-privacy.md>)

Tags: [completion](<https://devfeed.tech/tags/completion.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [electronic-health-records](<https://devfeed.tech/tags/electronic-health-records.md>), [health](<https://devfeed.tech/tags/health.md>), [risk](<https://devfeed.tech/tags/risk.md>), [stat-ap](<https://devfeed.tech/tags/stat-ap.md>), [synthesis](<https://devfeed.tech/tags/synthesis.md>)

### AI overview

EHRDiB unifies EHR visit completion and synthesis with a shared multimodal representation and an evidence-conditioned latent diffusion bridge, distinguishing original missingness from withheld inputs. Across MIMIC-IV and numerical and discretized Duke Health data, it had the smallest utility gaps in seven of eight completion comparisons. For synthesis, AUPRC gaps were 45.2% smaller on ten MIMIC-IV principal-diagnosis targets and 53.3% smaller on numerical Duke Health data than with the best comparator. Calibrated thresholding produced the lowest marginal distribution error across all three datasets, with coverage above 0.95 under both output rules. In MIMIC-IV synthesis, membership-inference AUROC rose from 0.516 to 0.598 as mean target evidence fraction increased from 0.20 to 0.50; utility gaps narrowed relative to the lowest-evidence setting but not monotonically, indicating a measured disclosure-risk tradeoff.

### Source excerpt

arXiv:2611.19546v1 Announce Type: new Abstract: Patient information can guide electronic health record (EHR) generation, but how its amount shapes predictive utility and disclosure risk remains poorly characterized. Here we present EHRDiB, which unifies visit completion and synthesis by distinguishing original missingness from withheld inputs and combining a shared multimodal representation with an evidence-conditioned latent diffusion bridge. We compared EHRDiB with three completion and eight synthesis methods on MIMIC-IV and numerical and discretized Duke Health representations. Evaluation covered downstream predictive utility, generative diversity, and distributional fidelity for both tasks, and disclosure risk for synthesis. EHRDiB achieved the smallest utility gaps from real-data-trained classifiers in seven of eight completion comparisons. In synthesis, AUPRC gaps were 45.2% smaller for ten MIMIC-IV principal-diagnosis targets and 53.3% smaller for numerical Duke Health data relative to the best comparator. Synthesis with calibrated thresholding yielded the lowest marginal distribution error across all three datasets, with Coverage above 0.95 for both output rules. Across twelve evidence-fraction intervals in MIMIC-IV synthesis, membership-inference AUROC increased from 0.516 to 0.598 as the mean target evidence fraction rose from 0.20 to 0.50; utility gaps narrowed relative to the lowest evidence configuration but did not decrease monotonically. Our study provides a unified framework and empirical foundation for EHR generation that considers both its utility for clinical analysis and its information disclosure risk.

## What can you safely tell AI at work?

DevFeed: [What can you safely tell AI at work?](<https://devfeed.tech/articles/what-can-you-safely-tell-ai-at-work-136230.md>)

Original publisher: [Read original article](<https://www.computerworld.com/article/4244131/what-can-you-safely-tell-ai-at-work.html>)

Author: Pat Brans

Published: 2026-11-16T12:04:12Z

Content type: news

Language: en

Sources: [Computerworld](<https://devfeed.tech/sources/computerworld.md>)

Topics: [shadow AI](<https://devfeed.tech/topics/shadow-ai.md>), [pii](<https://devfeed.tech/topics/pii.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [Access Control](<https://devfeed.tech/topics/access-control.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-at-work](<https://devfeed.tech/tags/ai-at-work.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [artificial-intelligence-generative-ai-it-leadership-it-strategy-laws-and-regulations-privacy-s](<https://devfeed.tech/tags/artificial-intelligence-generative-ai-it-leadership-it-strategy-laws-and-regulations-privacy-s.md>), [code](<https://devfeed.tech/tags/code.md>), [compliance](<https://devfeed.tech/tags/compliance.md>), [data-protection](<https://devfeed.tech/tags/data-protection.md>), [edition-us](<https://devfeed.tech/tags/edition-us.md>), [error-messages](<https://devfeed.tech/tags/error-messages.md>), [generative-ai](<https://devfeed.tech/tags/generative-ai.md>), [governance](<https://devfeed.tech/tags/governance.md>), [it-leadership](<https://devfeed.tech/tags/it-leadership.md>), [it-strategy](<https://devfeed.tech/tags/it-strategy.md>), [language-en](<https://devfeed.tech/tags/language-en.md>), [laws-and-regulations](<https://devfeed.tech/tags/laws-and-regulations.md>), [logs](<https://devfeed.tech/tags/logs.md>), [personally-identifiable-information](<https://devfeed.tech/tags/personally-identifiable-information.md>), [privacy](<https://devfeed.tech/tags/privacy.md>), [protection](<https://devfeed.tech/tags/protection.md>), [secrets](<https://devfeed.tech/tags/secrets.md>), [security](<https://devfeed.tech/tags/security.md>), [storytype-feature](<https://devfeed.tech/tags/storytype-feature.md>)

### AI overview

Troubleshooting details can expose sensitive information even after names are removed: service names, file paths, request sequences, customer numbers, meeting context, screenshots, and hidden spreadsheet columns may still identify people or reveal business processes. Sawant recommends reproducing issues with fictional data, simplifying until outsiders could not infer proprietary details, and using internal resources when safe sharing requires too much context. An approved AI tool still has limits; employees need to know what data it may handle and how inputs are stored, used, and accessed. Employers should clarify permitted accounts, tasks, and data through contracts, safeguards, examples, training, access controls, and review. The article also distinguishes confidentiality from policy compliance: using an unapproved tool may violate instructions even without disclosing client information. Asking in writing can document guidance but does not guarantee protection. The excerpt stresses practical support, escalation routes, monitoring, and updating controls as tools gain AI features.

### Source excerpt

When Bhaskar Bharat Sawant troubleshoots a software problem, an error message often provides the first clue. It can also present a dilemma: how much of that message can he safely share with AI tools? Sawant, a lead engineer and solution architect at building-products manufacturer Cornerstone Building Brands, works with legacy applications and newer technologies. GenAI can help explain an error or suggest a way forward. But the input it needs to understand the problem may expose personally identifiable information, a customer's competitive secrets, or his own company's proprietary information. "Logs or error messages can reveal internal service names, configuration details, or business processes," Sawant said. "You have to protect that information whether it belongs to your employer or a customer." A customer's name is an obvious identifier. A method name, file path, or sequence of requests between systems is less conspicuous. Together, those details can expose how an application works or how a business operates -- even when no individual can be identified. Sawant's approach is to construct a smaller, fictional version of the problem. "I create a small code example with made-up data that reproduces the issue," he said. "Then I ask myself whether someone outside the company could piece together proprietary information from it. If they could, I simplify it further or keep it out of the AI tool." That process takes judgment and effort before AI can save him any time. It also exposes the difficulty behind a simple workplace instruction: don't share confidential or sensitive information with generative AI. Employees must translate that warning into decisions about actual work, often while trying to meet a deadline. What remains after you remove the names Employees can recognize the need to protect information and still overlook what they are sharing. Julian Gage, who provides fractional data protection officer services through Engage Compliance, recalls advising a health t

## Everything you always wanted to know about Secure Exclaves but were afraid to ask

DevFeed: [Everything you always wanted to know about Secure Exclaves but were afraid to ask](<https://devfeed.tech/articles/everything-you-always-wanted-to-know-about-secure-exclaves-but-were-afraid-to-ask-136233.md>)

Original publisher: [Read original article](<https://www.computerworld.com/article/4249124/everything-you-always-wanted-to-know-about-secure-exclaves-but-were-afraid-to-ask.html>)

Author: Jonny Evans

Published: 2026-11-16T11:50:00Z

Content type: article

Language: en

Sources: [Computerworld](<https://devfeed.tech/sources/computerworld.md>)

Topics: [Protection](<https://devfeed.tech/topics/protection.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ambient](<https://devfeed.tech/tags/ambient.md>), [apple](<https://devfeed.tech/tags/apple.md>), [apple-privacy-security-vendors-and-providers](<https://devfeed.tech/tags/apple-privacy-security-vendors-and-providers.md>), [apple-silicon](<https://devfeed.tech/tags/apple-silicon.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [camera](<https://devfeed.tech/tags/camera.md>), [edition-us](<https://devfeed.tech/tags/edition-us.md>), [governance](<https://devfeed.tech/tags/governance.md>), [homeview](<https://devfeed.tech/tags/homeview.md>), [language-en](<https://devfeed.tech/tags/language-en.md>), [metadata](<https://devfeed.tech/tags/metadata.md>), [note](<https://devfeed.tech/tags/note.md>), [privacy](<https://devfeed.tech/tags/privacy.md>), [secure-exclaves](<https://devfeed.tech/tags/secure-exclaves.md>), [security](<https://devfeed.tech/tags/security.md>), [security-research](<https://devfeed.tech/tags/security-research.md>), [siri](<https://devfeed.tech/tags/siri.md>), [siri-recap](<https://devfeed.tech/tags/siri-recap.md>), [stored](<https://devfeed.tech/tags/stored.md>), [storytype-newsanalysis](<https://devfeed.tech/tags/storytype-newsanalysis.md>), [stream](<https://devfeed.tech/tags/stream.md>), [summary](<https://devfeed.tech/tags/summary.md>), [surveillance-technologies](<https://devfeed.tech/tags/surveillance-technologies.md>), [vendors-and-providers](<https://devfeed.tech/tags/vendors-and-providers.md>)

### AI overview

Apple's Secure Exclaves process ambient camera and microphone data in isolated runtimes, sharing only audited, task-specific metadata with the main operating system and discarding raw data after processing. The architecture uses an Apple-customized L4 microkernel and Secure Page Table Monitor for memory isolation. In ambient mode, only selected information reaches the system; in active uses such as Siri conversations or FaceTime, data may be released to the OS with a secure indicator. For HomeView person recognition, separate camera and audio exclaves use sensor input to identify a person, while the OS receives metadata such as identity, position, and confidence. Siri Recap and Live Rewind require opt-in. These protections and Apple's claims do not establish that other manufacturers offer similar safeguards, and the article notes that summaries remain accessible and the system is not claimed to be perfect.

### Source excerpt

We were first publicly introduced to Secure Exclaves when Apple introduced Siri Recap with watchOS 26. Now a recently published Apple security research paper sees them discussed as a protection for person recognition in HomeView. They are part of a new security architecture designed to protect user privacy even while computing becomes more ambient, smarter, and more pervasive. While it's good that Apple is thinking about such things, there is no guarantee at all that other companies, or even governments, will take steps to protect people's privacy as what are effectively surveillance technologies go mainstream. So, what are Secure Exclaves? Secure Exclaves are compartmentalized runtimes that make up a new security architecture in Apple silicon. The system uses an Apple-customized L4 microkernel, originally built for deployments like the Secure Enclave Processor. This provides a highly isolated environment in which to process camera and microphone data in such a way that only specific metadata is shared with the main operating system. The idea behind this is that this information is not stored, shared, or digitized in the conventional sense, even while the data itself can be processed. All that is stored is the condensed text and summary. One way to think about this is that even if you end up with a summarized note concerning the meeting, the raw data of the meeting itself is never stored, which means -- Apple claims -- it can't be accessed by your devices, Apple, or anyone else. (Though the summary text that is created can be accessed.) Overall, it's an architecture that lets AI work with data on the fly, working only with the real-life data stream most of the time. There are two primary modes for the tech: ambient and active. When in ambient mode, only specific information can be shared with the rest of the system. In active mode, such as a Siri conversation or a FaceTime call, data can be released to the main OS, with a secure indicator (usually a physical light) te

## How sensitive data from Centrelink and other Australian agencies may have been fed to Palantir software

DevFeed: [How sensitive data from Centrelink and other Australian agencies may have been fed to Palantir software](<https://devfeed.tech/articles/how-sensitive-data-from-centrelink-and-other-australian-agencies-may-have-been-fed-to-palantir-software-134676.md>)

Original publisher: [Read original article](<https://www.theguardian.com/technology/2026/nov/16/personal-data-centrelink-ato-ndis-palantir>)

Author: Josh Taylor Technology reporter

Published: 2026-11-15T21:06:03Z

Content type: news

Language: en

Sources: [The Guardian | Technology](<https://devfeed.tech/sources/the-guardian-technology.md>)

Topics: [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [Software](<https://devfeed.tech/topics/software.md>), [data loss prevention](<https://devfeed.tech/topics/data-loss-prevention.md>)

Tags: [agencies](<https://devfeed.tech/tags/agencies.md>), [australia](<https://devfeed.tech/tags/australia.md>), [australia-news](<https://devfeed.tech/tags/australia-news.md>), [australian-politics](<https://devfeed.tech/tags/australian-politics.md>), [fraud](<https://devfeed.tech/tags/fraud.md>), [information](<https://devfeed.tech/tags/information.md>), [palantir](<https://devfeed.tech/tags/palantir.md>), [palantir-australia-news-technology-australian-politics-peter-thiel](<https://devfeed.tech/tags/palantir-australia-news-technology-australian-politics-peter-thiel.md>), [personal-information](<https://devfeed.tech/tags/personal-information.md>), [peter-thiel](<https://devfeed.tech/tags/peter-thiel.md>), [sensitive-data](<https://devfeed.tech/tags/sensitive-data.md>), [software](<https://devfeed.tech/tags/software.md>), [surveillance](<https://devfeed.tech/tags/surveillance.md>), [technology](<https://devfeed.tech/tags/technology.md>)

### AI overview

Freedom of information documents show at least 10 federal agencies share sensitive data with the Australian Criminal Intelligence Commission's fraud fusion taskforce, where it could be analyzed using Palantir software. Services Australia authorized sharing information including Centrelink claims, income and assets, bank details and surveillance material; the ATO reported 42 disclosures to the ACIC, while tax file numbers cannot legally be provided. The ACIC described Palantir as one of several tools used for specific cases, but the excerpt does not establish whether particular shared data was ingested or whether supplying agencies were told. Documents also show a planned ethics and human oversight framework was replaced by an ethics guide after the framework was deemed too difficult to implement; a taskforce performance evaluation framework was not developed. The excerpt reports that the ACIC is not subject to the Privacy Act, though its protocols say it follows Australian privacy principles where reasonably consistent with its functions.

### Source excerpt

Exclusive: Personal information of Centrelink clients and childcare workers is among data shared with ACIC, which uses US tech company's software The personal information of Centrelink customers, taxpayers and childcare workers is among a trove of sensitive data that may be ending up in Palantir's surveillance software in Australia, without the departments responsible for the information even knowing about it. As Guardian Australia reported in September, Palantir's software is used by the Australian Criminal Intelligence Commission (ACIC) as part of the multi-agency fraud fusion taskforce, which was established in 2022 with the goal of targeting fraud in government payments from the national disability insurance scheme, childcare and welfare. Continue reading...

## TSMC employee data breach linked to cloud provider

DevFeed: [TSMC employee data breach linked to cloud provider](<https://devfeed.tech/articles/tsmc-employee-data-breach-linked-to-cloud-provider-134458.md>)

Original publisher: [Read original article](<https://www.bleepingcomputer.com/news/security/tsmc-employee-data-breach-linked-to-cloud-provider/>)

Author: Bill Toulas

Published: 2026-11-15T15:23:48Z

Content type: news

Language: en

Sources: [BleepingComputer](<https://devfeed.tech/sources/bleepingcomputer.md>)

Topics: [tsmc](<https://devfeed.tech/topics/tsmc.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [data](<https://devfeed.tech/topics/data.md>)

Tags: [000](<https://devfeed.tech/tags/000.md>), [3](<https://devfeed.tech/tags/3.md>), [actor](<https://devfeed.tech/tags/actor.md>), [auction](<https://devfeed.tech/tags/auction.md>), [breach](<https://devfeed.tech/tags/breach.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-provider](<https://devfeed.tech/tags/cloud-provider.md>), [cloud-security](<https://devfeed.tech/tags/cloud-security.md>), [company](<https://devfeed.tech/tags/company.md>), [computer-help](<https://devfeed.tech/tags/computer-help.md>), [computer-security](<https://devfeed.tech/tags/computer-security.md>), [computers](<https://devfeed.tech/tags/computers.md>), [data](<https://devfeed.tech/tags/data.md>), [data-breach](<https://devfeed.tech/tags/data-breach.md>), [data-leak](<https://devfeed.tech/tags/data-leak.md>), [employee](<https://devfeed.tech/tags/employee.md>), [information](<https://devfeed.tech/tags/information.md>), [infosec](<https://devfeed.tech/tags/infosec.md>), [infosec-computer-security](<https://devfeed.tech/tags/infosec-computer-security.md>), [linux](<https://devfeed.tech/tags/linux.md>), [mac](<https://devfeed.tech/tags/mac.md>), [malware](<https://devfeed.tech/tags/malware.md>), [malware-removal](<https://devfeed.tech/tags/malware-removal.md>), [manufacturing](<https://devfeed.tech/tags/manufacturing.md>), [online](<https://devfeed.tech/tags/online.md>), [personal](<https://devfeed.tech/tags/personal.md>), [sale](<https://devfeed.tech/tags/sale.md>), [security](<https://devfeed.tech/tags/security.md>), [semiconductor](<https://devfeed.tech/tags/semiconductor.md>), [semiconductor-manufacturing](<https://devfeed.tech/tags/semiconductor-manufacturing.md>), [spyware](<https://devfeed.tech/tags/spyware.md>), [support](<https://devfeed.tech/tags/support.md>), [taiwan](<https://devfeed.tech/tags/taiwan.md>), [tech-support](<https://devfeed.tech/tags/tech-support.md>), [technical-support](<https://devfeed.tech/tags/technical-support.md>), [threat-actor](<https://devfeed.tech/tags/threat-actor.md>), [tsmc](<https://devfeed.tech/tags/tsmc.md>), [virus](<https://devfeed.tech/tags/virus.md>), [virus-removal](<https://devfeed.tech/tags/virus-removal.md>), [windows](<https://devfeed.tech/tags/windows.md>)

### AI overview

TSMC confirmed that current and former employee data appeared online and matched information stored in a cloud service provider's systems. The company said it found no evidence that its own systems were compromised and that the incident did not affect operations or customers. It is notifying affected staff, providing support, and working with a third-party cybersecurity firm. A threat actor claimed a 6.9GB dataset held 46,000 employee records, posted a 499-row sample, and offered the data at auction; the record count and sensitive fields described in the post are the actor's claims. TSMC declined to comment on whether it had contacted or planned to pay the actor.

### Source excerpt

Taiwan Semiconductor Manufacturing Company (TSMC) confirmed that personal employee data has been leaked online after a threat actor put 6.9GB of information allegedly stolen from the company up for auction. [...]

## Logging AI Prompts and Responses in Java Without Leaking Sensitive Data

DevFeed: [Logging AI Prompts and Responses in Java Without Leaking Sensitive Data](<https://devfeed.tech/articles/logging-ai-prompts-and-responses-in-java-without-leaking-sensitive-data-133710.md>)

Original publisher: [Read original article](<https://www.sitepoint.com/logging-ai-prompts-and-responses-in-java-without-leaking-sensitive-data/>)

Author: TechPanda

Published: 2026-11-14T06:15:17Z

Content type: article

Language: en

Sources: [SitePoint](<https://devfeed.tech/sources/sitepoint.md>)

Topics: [API Logging](<https://devfeed.tech/topics/api-logging.md>), [data loss prevention](<https://devfeed.tech/topics/data-loss-prevention.md>), [Logging](<https://devfeed.tech/topics/logging.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [Java](<https://devfeed.tech/topics/java.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-prompts](<https://devfeed.tech/tags/ai-prompts.md>), [data](<https://devfeed.tech/tags/data.md>), [java](<https://devfeed.tech/tags/java.md>), [logging](<https://devfeed.tech/tags/logging.md>), [responses](<https://devfeed.tech/tags/responses.md>), [sensitive-data](<https://devfeed.tech/tags/sensitive-data.md>)

### AI overview

The article recommends logging metadata for every Java AI call while limiting prompt and response content to masked samples and failed calls. Route all content through one PiiMasker before it reaches any logger; use compiled, ordered regex rules for structured data such as emails, card numbers, phone numbers, and bearer tokens. Regex may miss names and sensitive details in free text, so combine it with entity recognition, known-value replacement, and field-level masking during prompt construction. For tracing repeated values, use keyed HMAC pseudonyms rather than unkeyed hashes. Write structured JSON records, keep content briefly, and restrict access. Assemble streamed responses before masking because sensitive values can span chunks; treat tool arguments and results as content too. The excerpt also advises testing masking against messy examples and guarding against raw payloads in debug logs, SDK exceptions, and wire-level HTTP logging.

### Source excerpt

null Continue reading Logging AI Prompts and Responses in Java Without Leaking Sensitive Data on SitePoint.

## The Green Agent POV: Formulating the Fix for an Internet-Facing Container with Sensitive Data Access

DevFeed: [The Green Agent POV: Formulating the Fix for an Internet-Facing Container with Sensitive Data Access](<https://devfeed.tech/articles/the-green-agent-pov-formulating-the-fix-for-an-internet-facing-container-with-sensitive-data-access-133499.md>)

Original publisher: [Read original article](<https://www.wiz.io/blog/inside-green-agent-investigation>)

Author: Green Agent

Published: 2026-11-13T17:05:02Z

Content type: article

Language: en

Sources: [Wiz](<https://devfeed.tech/sources/wiz-blog-rss-feed.md>)

Topics: [Security](<https://devfeed.tech/topics/security.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [network security](<https://devfeed.tech/topics/network-security.md>), [container image compliance](<https://devfeed.tech/topics/container-image-compliance.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [zero downtime deployment](<https://devfeed.tech/topics/zero-downtime-deployment.md>)

Tags: [000](<https://devfeed.tech/tags/000.md>), [2](<https://devfeed.tech/tags/2.md>), [2026](<https://devfeed.tech/tags/2026.md>), [22](<https://devfeed.tech/tags/22.md>), [3](<https://devfeed.tech/tags/3.md>), [agent](<https://devfeed.tech/tags/agent.md>), [business](<https://devfeed.tech/tags/business.md>), [business-continuity](<https://devfeed.tech/tags/business-continuity.md>), [container](<https://devfeed.tech/tags/container.md>), [data](<https://devfeed.tech/tags/data.md>), [green](<https://devfeed.tech/tags/green.md>), [internet](<https://devfeed.tech/tags/internet.md>), [merge](<https://devfeed.tech/tags/merge.md>), [plan](<https://devfeed.tech/tags/plan.md>), [pov](<https://devfeed.tech/tags/pov.md>), [pull-requests](<https://devfeed.tech/tags/pull-requests.md>), [remediation](<https://devfeed.tech/tags/remediation.md>), [security](<https://devfeed.tech/tags/security.md>), [sensitive-data](<https://devfeed.tech/tags/sensitive-data.md>), [with](<https://devfeed.tech/tags/with.md>), [wiz-agents](<https://devfeed.tech/tags/wiz-agents.md>)

### AI overview

Green Agent investigated a publicly reachable Kubernetes workload whose actively loaded Node.js Brotli vulnerability created a path to an IAM role and more than 5,000 sensitive S3 records. It traced the internet route and confirmed unauthenticated endpoints with no WAF or firewall, while determining that removing ingress would take down an intentionally public production service. Since vendor patches were available but image deployment would take time, it identified immediate cloud mitigations, including disabling service account token automount and narrowing IAM permissions, alongside permanent code and Terraform fixes. It traced relevant files to their repositories and owners, then assigned mitigation and fix plans. The account says the agent completed this work in minutes; it provides no independent evaluation of that claim.

### Source excerpt

See how the Green Agent balanced security urgency with business continuity to turn a toxic combination into an executable remediation plan and ready-to-merge pull requests.

## ZDI-26-993: ATEN eco DC chartDefine SQL Injection Information Disclosure Vulnerability

DevFeed: [ZDI-26-993: ATEN eco DC chartDefine SQL Injection Information Disclosure Vulnerability](<https://devfeed.tech/articles/zdi-26-993-aten-eco-dc-chartdefine-sql-injection-information-disclosure-vulnerability-133818.md>)

Original publisher: [Read original article](<http://www.zerodayinitiative.com/advisories/ZDI-26-993/>)

Published: 2026-11-13T06:00:00Z

Content type: news

Language: en

Sources: [Zero Day Initiative](<https://devfeed.tech/sources/zero-day-initiative.md>)

Topics: [Exploit](<https://devfeed.tech/topics/exploit.md>), [Vulnerabilities](<https://devfeed.tech/topics/vulnerabilities.md>), [SQL](<https://devfeed.tech/topics/sql.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>)

Tags: [sql-injection](<https://devfeed.tech/tags/sql-injection.md>), [vulnerability](<https://devfeed.tech/tags/vulnerability.md>)

### AI overview

An authenticated remote attacker can exploit improper validation of color references used to create charts in ATEN eco DC, where the supplied string is incorporated into SQL queries. This can disclose stored credentials and lead to further compromise. The issue is fixed in eco DC 1.2.119.

### Source excerpt

This vulnerability allows remote attackers to disclose sensitive information on affected installations of ATEN eco DC. Authentication is required to exploit this vulnerability. The ZDI has assigned a CVSS rating of 2.7. The following CVEs are assigned: CVE-2026-126965.

[Next page](<https://devfeed.tech/topics/sensitive-data.md?cursor=WyIyMDI2LTExLTEzVDA2OjAwOjAwKzAwOjAwIiwgIjMxODhmYWVmLThkZjgtNDg2ZC05MDAzLWZhZDM3ODlmNWM5YyJd>)