# Risk

A measure of how extensively an entity is threatened by a potential circumstance or event, based on adverse impact and likelihood of occurrence.

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

## The Attack Path Problem: How Complexity Hides Risk in Financial Services

DevFeed: [The Attack Path Problem: How Complexity Hides Risk in Financial Services](<https://devfeed.tech/articles/the-attack-path-problem-how-complexity-hides-risk-in-financial-services-136614.md>)

Original publisher: [Read original article](<https://www.netspi.com/blog/executive-blog/penetration-testing-as-a-service/the-attack-path-problem-how-complexity-hides-risk-in-financial-services/>)

Author: Kyle Grauberger

Published: 2026-11-16T19:18:38Z

Content type: article

Language: en

Sources: [NetSPI](<https://devfeed.tech/sources/netspi.md>)

Topics: [Security](<https://devfeed.tech/topics/security.md>), [Risk](<https://devfeed.tech/topics/risk.md>), [Automation](<https://devfeed.tech/topics/automation.md>), [Authorization](<https://devfeed.tech/topics/authorization.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-driven-attacks](<https://devfeed.tech/tags/ai-driven-attacks.md>), [api](<https://devfeed.tech/tags/api.md>), [apis](<https://devfeed.tech/tags/apis.md>), [attack-paths](<https://devfeed.tech/tags/attack-paths.md>), [attack-surface](<https://devfeed.tech/tags/attack-surface.md>), [automation](<https://devfeed.tech/tags/automation.md>), [autonomous-pentesting](<https://devfeed.tech/tags/autonomous-pentesting.md>), [banking](<https://devfeed.tech/tags/banking.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [complexity](<https://devfeed.tech/tags/complexity.md>), [compliance](<https://devfeed.tech/tags/compliance.md>), [continuous](<https://devfeed.tech/tags/continuous.md>), [continuous-penetration-testing](<https://devfeed.tech/tags/continuous-penetration-testing.md>), [continuous-pentesting](<https://devfeed.tech/tags/continuous-pentesting.md>), [financial-services](<https://devfeed.tech/tags/financial-services.md>), [penetration-testing](<https://devfeed.tech/tags/penetration-testing.md>), [risk](<https://devfeed.tech/tags/risk.md>)

### AI overview

Financial institutions face risk from how findings connect across legacy platforms, cloud apps, APIs, partners, and service accounts; severity lists alone do not show whether weaknesses could enable outcomes such as account takeover or fraudulent transfers. The article says AI can scale asset discovery, exploration, evidence gathering, validation, and retesting, while continuous testing better tracks frequent releases and cloud changes than quarterly or annual testing alone. It cautions that fully autonomous pentesting still struggles to connect findings across systems and assess access in regulated environments, potentially creating gaps or noise. Its recommended current approach pairs AI agents with experienced testers who direct work, investigate unusual paths, validate impact, and link findings into plausible attack paths. The article presents NetSPI+Synack's service as combining agentic AI and human expertise for continuous or on-demand testing; this is the provider's claim.

### Source excerpt

If you work in financial services, you know the scale of findings isn't the hardest part. How the findings connect is what really matters. Learn how to pressure-test your security approach against modern AI attacks to your actual environment. The post The Attack Path Problem: How Complexity Hides Risk in Financial Services appeared first on NetSPI.

## Rare Event Estimation via Iterative Unalignment

DevFeed: [Rare Event Estimation via Iterative Unalignment](<https://devfeed.tech/articles/rare-event-estimation-via-iterative-unalignment-131971.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2609.24969>)

Author: Hanming Yang, Daksh Mittal, Jing Dong, Hongseok Namkoong

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

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

Topics: [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [Risk](<https://devfeed.tech/topics/risk.md>)

Tags: [adaptive](<https://devfeed.tech/tags/adaptive.md>), [agent](<https://devfeed.tech/tags/agent.md>), [autonomy](<https://devfeed.tech/tags/autonomy.md>), [conditional](<https://devfeed.tech/tags/conditional.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [estimator](<https://devfeed.tech/tags/estimator.md>), [events](<https://devfeed.tech/tags/events.md>), [gradient](<https://devfeed.tech/tags/gradient.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [model](<https://devfeed.tech/tags/model.md>), [models](<https://devfeed.tech/tags/models.md>), [monte-carlo](<https://devfeed.tech/tags/monte-carlo.md>), [probability](<https://devfeed.tech/tags/probability.md>), [risk](<https://devfeed.tech/tags/risk.md>), [sampling](<https://devfeed.tech/tags/sampling.md>), [unalignment](<https://devfeed.tech/tags/unalignment.md>)

### AI overview

The paper estimates rare events in stochastic agent outputs by using importance sampling (IS) with a proposal model created by perturbing the original language model's weights. Because the proposal is differentiably parameterized, the method searches weight space with gradients; its objective combines a differentiable event-amplification surrogate with adaptive regularization to balance amplification and estimator stability. The authors evaluate it on approximately 120M- and 2.6B-parameter models across three event families and more than 300 events, including probabilities as low as 10⁻⁹, using independently estimated reference probabilities. In the most verifiable settings, for events rarer than 10⁻⁷, they report average compute-weighted efficiency gains above 800x versus naive Monte Carlo.

### Source excerpt

arXiv:2609.24969v2 Announce Type: replace-cross Abstract: As agents are deployed with increased autonomy, even extremely rare events along their stochastic output trajectories can occur and prove catastrophic. Safe deployment therefore does not depend on whether these events can occur, but on how often they might. We study the problem of estimating the probability of rare events that arise from stochastic variation in the agent's own actions. Estimating this type of risk requires searching over the combinatorially vast space of trajectories. Naive Monte Carlo is computationally prohibitive in this regime, and constructing effective importance sampling (IS) proposals requires coordinated changes to a context-dependent chain of conditional distributions. We develop a new IS method that perturbs the original model's weights to construct the proposal. The proposal is itself a differentiably parameterized language model, enabling gradient-based search over weight space. We formulate an objective that combines a differentiable surrogate for event amplification and an adaptive regularization scheme that dynamically balances amplification against estimator stability. We evaluate our approach on $\sim$120M and $\sim$2.6B models across three event families spanning 300+ rare events as rare as $10^{-9}$, all with independently estimated reference probabilities. In our most verifiable settings, we observe that our IS estimator achieves over $800\times$ compute-weighted efficiency gains on average over naive Monte Carlo for events with probabilities lower than $10^{-7}$. Our implementation is available at https://github.com/namkoong-lab/iterative-unalignment.

## How Fake Job Interviews Helped Hackers Pull Off a $6.3 Billion Crypto Heist

DevFeed: [How Fake Job Interviews Helped Hackers Pull Off a $6.3 Billion Crypto Heist](<https://devfeed.tech/articles/how-fake-job-interviews-helped-hackers-pull-off-a-6-3-billion-crypto-heist-128450.md>)

Original publisher: [Read original article](<https://hackernoon.com/how-fake-job-interviews-helped-hackers-pull-off-a-$63-billion-crypto-heist?source=rss>)

Author: Samuel Oladipupo

Published: 2026-11-11T07:35:09Z

Content type: article

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [fake account fraud](<https://devfeed.tech/topics/fake-account-fraud.md>), [Risk](<https://devfeed.tech/topics/risk.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [binance](<https://devfeed.tech/tags/binance.md>), [browser](<https://devfeed.tech/tags/browser.md>), [crypto](<https://devfeed.tech/tags/crypto.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [engineers](<https://devfeed.tech/tags/engineers.md>), [fake-job-interviews](<https://devfeed.tech/tags/fake-job-interviews.md>), [hackers](<https://devfeed.tech/tags/hackers.md>), [job-interview](<https://devfeed.tech/tags/job-interview.md>), [malware](<https://devfeed.tech/tags/malware.md>), [malware-attacks](<https://devfeed.tech/tags/malware-attacks.md>), [online-safety](<https://devfeed.tech/tags/online-safety.md>), [run-untrusted-code](<https://devfeed.tech/tags/run-untrusted-code.md>), [sandbox](<https://devfeed.tech/tags/sandbox.md>), [social-engineering](<https://devfeed.tech/tags/social-engineering.md>), [social-engineering-attacks](<https://devfeed.tech/tags/social-engineering-attacks.md>)

### AI overview

Binance says hackers used fake job interviews to compromise an engineer: after building trust through a legitimate job site and social media accounts, the impostors sent malware disguised as an assignment. The attackers then used the engineer's computer and an internal AI tool to take over two other engineers' laptops. The excerpt recommends that candidates avoid installing recruiter-provided software or running test files on work devices, verify employers through contact details found independently, and use a separate device if a coding task must be run. For employers, it suggests browser-based or sandboxed take-home tasks, written clarity about installation requests, and a verifiable named contact. The account is attributed to Binance; the excerpt does not independently establish the reported loss figure or provide further technical detail.

### Source excerpt

Binance says fake job interviews helped hackers steal a record $6.3bn. Here is how it worked, and what to do differently.

## Entropic Risk-Aware Monte Carlo Tree Search

DevFeed: [Entropic Risk-Aware Monte Carlo Tree Search](<https://devfeed.tech/articles/entropic-risk-aware-monte-carlo-tree-search-128499.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2601.17667>)

Author: Pedro P. Santos, Jacopo Silvestrin, Manuel Lopes, Alberto Sardinha, Francisco S. Melo

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

Content type: article

Language: en

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

Topics: [monte carlo tree search](<https://devfeed.tech/topics/monte-carlo-tree-search.md>), [Risk](<https://devfeed.tech/topics/risk.md>)

Tags: [algorithm](<https://devfeed.tech/tags/algorithm.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [dynamic-programming](<https://devfeed.tech/tags/dynamic-programming.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [markov](<https://devfeed.tech/tags/markov.md>), [monte-carlo-tree-search](<https://devfeed.tech/tags/monte-carlo-tree-search.md>), [risk](<https://devfeed.tech/tags/risk.md>)

### AI overview

The paper proposes an upper confidence bound-based Monte Carlo tree search algorithm for risk-aware Markov decision processes with entropic risk measure objectives. Its non-asymptotic analysis shows that the empirical risk measure at the root converges to the optimal value and establishes polynomial regret concentration. The method uses dynamic programming formulations for these objectives; illustrative experiments show a tradeoff between risk-neutral and risk-averse behavior.

### Source excerpt

arXiv:2601.17667v3 Announce Type: replace Abstract: We propose a provably correct Monte Carlo tree search (MCTS) algorithm for solving risk-aware Markov decision processes (MDPs) with entropic risk measure (ERM) objectives. We provide a non-asymptotic analysis of our proposed algorithm, showing that the algorithm: (i) is correct in the sense that the empirical ERM obtained at the root node converges to the optimal ERM; and (ii) enjoys polynomial regret concentration. Our algorithm successfully exploits the dynamic programming formulations for solving risk-aware MDPs with ERM objectives introduced by previous works in the context of an upper confidence bound-based tree search algorithm. Finally, we provide a set of illustrative experiments to evaluate our risk-aware MCTS method, showing it successfully trades off between risk-neutral and risk-averse behavior.

## How do we auto-approve and merge 60% of PRs?

DevFeed: [How do we auto-approve and merge 60% of PRs?](<https://devfeed.tech/articles/how-do-we-auto-approve-and-merge-60-of-prs-122498.md>)

Original publisher: [Read original article](<https://swizec.com/blog/how-do-we-auto-approve-and-merge-60p-of-prs>)

Author: hi@swizec.com (Swizec Teller)

Published: 2026-11-07T00:00:00Z

Content type: opinion

Language: en

Sources: [Swizec Teller](<https://devfeed.tech/sources/swizec-teller.md>)

Topics: [Code review](<https://devfeed.tech/topics/code-review.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>), [engineering-culture](<https://devfeed.tech/topics/engineering-culture.md>), [Risk](<https://devfeed.tech/topics/risk.md>), [ai code review](<https://devfeed.tech/topics/ai-code-review.md>), [human review](<https://devfeed.tech/topics/human-review.md>), [Continuous integration](<https://devfeed.tech/topics/continuous-integration.md>), [developer-productivity](<https://devfeed.tech/topics/developer-productivity.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [human centric design](<https://devfeed.tech/topics/human-centric-design.md>), [UI/UX Design](<https://devfeed.tech/topics/ui-ux-design.md>), [API Governance](<https://devfeed.tech/topics/api-governance.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [3](<https://devfeed.tech/tags/3.md>), [about](<https://devfeed.tech/tags/about.md>), [after](<https://devfeed.tech/tags/after.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-testing](<https://devfeed.tech/tags/agent-testing.md>), [agents](<https://devfeed.tech/tags/agents.md>), [all](<https://devfeed.tech/tags/all.md>), [also](<https://devfeed.tech/tags/also.md>), [api](<https://devfeed.tech/tags/api.md>), [apis](<https://devfeed.tech/tags/apis.md>), [but](<https://devfeed.tech/tags/but.md>), [can](<https://devfeed.tech/tags/can.md>), [code](<https://devfeed.tech/tags/code.md>), [code-review](<https://devfeed.tech/tags/code-review.md>), [getting](<https://devfeed.tech/tags/getting.md>), [i](<https://devfeed.tech/tags/i.md>), [it-s](<https://devfeed.tech/tags/it-s.md>), [merge](<https://devfeed.tech/tags/merge.md>), [review](<https://devfeed.tech/tags/review.md>)

### AI overview

The author argues that reaching 60% auto-merged PRs requires shifting human review toward design and user outcomes, while classifying and automating routine code changes. Their team discusses architecture and product direction before implementation, then tests working features and reviews UX; code review can follow once behavior is verified. Their current bot excludes migrations, public interfaces, finance, and medium- or high-risk changes, while merging approved medium-risk PRs without comments. Proposed changes include finer risk classification, detecting destructive migrations and API contract changes, auto-fixing simple issues, and end-to-end tests for critical flows. A staged process would review direction before coding, use agents and checks during coding, and evaluate behavior afterward, with some noncritical checks potentially deferred for quick follow-up fixes. The author presents these as plans and ideas, not proven results.

### Source excerpt

I'm still drowning in code review. Getting to ~16% auto-approved and merged PRs helped but it's not enough. How can we do 60%?

## Volatility-Aware Extreme Event Detection in High-Frequency Financial Markets

DevFeed: [Volatility-Aware Extreme Event Detection in High-Frequency Financial Markets](<https://devfeed.tech/articles/volatility-aware-extreme-event-detection-in-high-frequency-financial-markets-116729.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2607.17555>)

Author: Maorufa Zaman, Haris Md Sahed

Published: 2026-11-04T05:00:00Z

Content type: article

Language: en

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

Topics: [Finance](<https://devfeed.tech/topics/finance.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [clustering](<https://devfeed.tech/topics/clustering.md>), [outlier-detection](<https://devfeed.tech/topics/outlier-detection.md>), [Risk](<https://devfeed.tech/topics/risk.md>)

Tags: [bitcoin](<https://devfeed.tech/tags/bitcoin.md>), [clustering](<https://devfeed.tech/tags/clustering.md>), [cross-validation](<https://devfeed.tech/tags/cross-validation.md>), [cryptocurrency](<https://devfeed.tech/tags/cryptocurrency.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [detection](<https://devfeed.tech/tags/detection.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [event](<https://devfeed.tech/tags/event.md>), [financial](<https://devfeed.tech/tags/financial.md>), [financial-markets](<https://devfeed.tech/tags/financial-markets.md>), [labeling](<https://devfeed.tech/tags/labeling.md>), [learning](<https://devfeed.tech/tags/learning.md>), [limit-order-book](<https://devfeed.tech/tags/limit-order-book.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [markets](<https://devfeed.tech/tags/markets.md>), [model](<https://devfeed.tech/tags/model.md>), [time-series](<https://devfeed.tech/tags/time-series.md>)

### AI overview

Using high-frequency Bitcoin limit order book data, the study reports that labeling events to include both large future returns and high-volatility regimes improved XGBoost detection of rare events: PR-AUC was approximately 0.40 versus 0.06 for the baseline, using time-series cross-validation and imbalance-aware evaluation. The authors attribute the gain to target design informed by volatility clustering and note that it can matter more than model complexity. The supplied abstract does not describe further methods or limitations.

### Source excerpt

arXiv:2607.17555v2 Announce Type: replace Abstract: Predicting extreme price movements in high-frequency financial markets is a challenging task due to non-stationarity, heavy-tailed return distributions, and severe class imbalance. In particular, rare but impactful events are often difficult to detect using conventional modeling approaches, which typically treat extreme movements as isolated observations. This study proposes a volatility-aware approach for extreme event detection using high-frequency Bitcoin limit order book (LOB) data. Motivated by empirical evidence of volatility clustering, the target formulation is extended to incorporate both large future returns and high-volatility regimes. This redefinition increases the proportion of informative samples and aligns the learning objective with the underlying market dynamics. Using a tree-based model (XGBoost) with time-series cross-validation and imbalance-aware evaluation, the proposed method achieves a Precision-Recall AUC of approximately 0.40, significantly outperforming the baseline formulation with a PR-AUC of around 0.06. This represents more than a sixfold improvement in detecting rare events. The results highlight that target design plays a critical role in financial machine learning, often exceeding the impact of model complexity. By incorporating volatility structure into the labeling process, the proposed approach provides a more effective and realistic framework for extreme event detection in high-frequency cryptocurrency markets.

## TAPES: Task-Aligned Joint Predictive Pretraining for Transaction Event Sequences

DevFeed: [TAPES: Task-Aligned Joint Predictive Pretraining for Transaction Event Sequences](<https://devfeed.tech/articles/tapes-task-aligned-joint-predictive-pretraining-for-transaction-event-sequences-116977.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.04747>)

Author: Ruobing Zuo, Weiyu Han, Huilin He, Haoyu Jiang, Ze Zhao, Jie Sheng, Xiao Tan, Shuai Chen, Tianyi Zhang

Published: 2026-11-04T05:00:00Z

Content type: article

Language: en

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

Topics: [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>), [Representation learning](<https://devfeed.tech/topics/representation-learning.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [Risk](<https://devfeed.tech/topics/risk.md>)

Tags: [ablation](<https://devfeed.tech/tags/ablation.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [event](<https://devfeed.tech/tags/event.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [financial](<https://devfeed.tech/tags/financial.md>), [financial-services](<https://devfeed.tech/tags/financial-services.md>), [online-commerce](<https://devfeed.tech/tags/online-commerce.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [pretraining](<https://devfeed.tech/tags/pretraining.md>), [risk](<https://devfeed.tech/tags/risk.md>), [sequences](<https://devfeed.tech/tags/sequences.md>), [target](<https://devfeed.tech/tags/target.md>), [tasks](<https://devfeed.tech/tags/tasks.md>), [transaction](<https://devfeed.tech/tags/transaction.md>), [utility](<https://devfeed.tech/tags/utility.md>)

### AI overview

TAPES selects transaction prediction targets by testing candidate sets in short joint pretraining trials and comparing the downstream utility of the resulting representations under matched budgets. It then reinitializes the encoder and jointly pretrains on the selected targets using a likelihood parameterized by a shared discrete latent variable, followed by supervised adaptation. The approach addresses the gap between target predictability and downstream usefulness, and models conditional co-occurrence among targets. The abstract reports better performance than representative baselines in a transaction-risk setting and across public financial benchmarks, with ablations supporting target selection and joint modeling; it provides no quantitative results.

### Source excerpt

arXiv:2611.04747v1 Announce Type: new Abstract: Transaction event sequences support diverse prediction tasks in online commerce and digital financial services. The same transaction records can yield prediction targets with different semantics, including values, changes, and relations to past events. These targets impose different constraints on history representations, but their predictability alone does not establish the downstream utility of a target set. Moreover, target-wise prediction losses do not explicitly model conditional co-occurrence among targets. To address these challenges, we propose TAPES, a task-aligned joint predictive pretraining framework for transaction sequences. TAPES constructs a finite collection of candidate target sets from admissible field-operator combinations. It selects a target set according to the downstream utility of representations learned through short joint pretraining trials under matched budgets. It then reinitializes the encoder and pretrains it on the selected targets through a joint likelihood parameterized by a shared discrete latent variable, before downstream supervised adaptation. Experiments in a real-world transaction-risk setting and on multiple public financial benchmarks spanning diverse downstream tasks demonstrate that TAPES outperforms representative baselines. Ablation studies further validate the contributions of predictive target selection and joint modeling.

## From Bug Report to Production: How I Fix Bugs with AI

DevFeed: [From Bug Report to Production: How I Fix Bugs with AI](<https://devfeed.tech/articles/from-bug-report-to-production-how-i-fix-bugs-with-ai-111025.md>)

Original publisher: [Read original article](<https://hackernoon.com/from-bug-report-to-production-how-i-fix-bugs-with-ai?source=rss>)

Author: inDrive.Tech

Published: 2026-11-01T21:40:15Z

Content type: article

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [bugfixes](<https://devfeed.tech/topics/bugfixes.md>), [Monitoring](<https://devfeed.tech/topics/monitoring.md>), [Static code analysis](<https://devfeed.tech/topics/static-code-analysis.md>), [engineering-culture](<https://devfeed.tech/topics/engineering-culture.md>), [Risk](<https://devfeed.tech/topics/risk.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [ai-coding-agents](<https://devfeed.tech/tags/ai-coding-agents.md>), [api-testing](<https://devfeed.tech/tags/api-testing.md>), [bugs](<https://devfeed.tech/tags/bugs.md>), [claude](<https://devfeed.tech/tags/claude.md>), [clusters](<https://devfeed.tech/tags/clusters.md>), [code](<https://devfeed.tech/tags/code.md>), [code-review](<https://devfeed.tech/tags/code-review.md>), [good-company](<https://devfeed.tech/tags/good-company.md>), [hackernoon-top-story](<https://devfeed.tech/tags/hackernoon-top-story.md>), [indrive](<https://devfeed.tech/tags/indrive.md>), [logs](<https://devfeed.tech/tags/logs.md>), [production](<https://devfeed.tech/tags/production.md>), [production-monitoring](<https://devfeed.tech/tags/production-monitoring.md>), [qa](<https://devfeed.tech/tags/qa.md>), [qa-engineer](<https://devfeed.tech/tags/qa-engineer.md>), [qa-engineering](<https://devfeed.tech/tags/qa-engineering.md>), [quality-assurance](<https://devfeed.tech/tags/quality-assurance.md>), [ship](<https://devfeed.tech/tags/ship.md>), [software-testing](<https://devfeed.tech/tags/software-testing.md>)

### AI overview

The article describes a three-level workflow in which a QA engineer uses Claude to investigate code, verify hypotheses with production logs and metrics, then prepare and ship small fixes through normal review and release checks. Bug reports include a proposed fix, regression criteria, risks, user impact, and code references, with confirmed findings distinguished from hypotheses. Production analysis can reveal misleading error counts: one week of data across 13 clusters showed about 1.86 million 500 errors, roughly 96.6% tied to client cancellations; the proposed handling change was an estimate, not a deployed result. A shipped idempotency-key fix validated requests before reserving keys and released them after failures; duplicate-key errors fell about 30%, with no noticeable response-time or other-error change and an assessed cache load of about 0.01%. The engineer retains responsibility for evidence, scope, tests, review, release, and post-release checks.

### Source excerpt

How a QA engineer at inDrive uses AI to investigate bugs, analyze production logs, write fixes, and ship changes to production.

## PHP2Shell Is Already Here: How the Infoblox Platform Helps Uncover and Mitigate the Risk

DevFeed: [PHP2Shell Is Already Here: How the Infoblox Platform Helps Uncover and Mitigate the Risk](<https://devfeed.tech/articles/php2shell-is-already-here-how-the-infoblox-platform-helps-uncover-and-mitigate-the-risk-107538.md>)

Original publisher: [Read original article](<https://www.infoblox.com/blog/security/php2shell-is-already-here-how-the-infoblox-platform-helps-uncover-and-mitigate-the-risk/>)

Author: Daniel Brody

Published: 2026-10-29T13:00:53Z

Content type: article

Language: en

Sources: [Infoblox](<https://devfeed.tech/sources/infoblox-blog.md>)

Topics: [infoblox](<https://devfeed.tech/topics/infoblox.md>), [known exploitable vulnerabilities](<https://devfeed.tech/topics/known-exploitable-vulnerabilities.md>), [observability](<https://devfeed.tech/topics/observability.md>), [Frontier AI](<https://devfeed.tech/topics/frontier-ai.md>), [Security research](<https://devfeed.tech/topics/security-research.md>), [cybersecurity and infrastructure security agency](<https://devfeed.tech/topics/cybersecurity-and-infrastructure-security-agency.md>), [Memory Safety](<https://devfeed.tech/topics/memory-safety.md>), [Risk](<https://devfeed.tech/topics/risk.md>), [Cloudflare](<https://devfeed.tech/topics/cloudflare.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-attack](<https://devfeed.tech/tags/ai-attack.md>), [ai-powered](<https://devfeed.tech/tags/ai-powered.md>), [anomaly-detection](<https://devfeed.tech/tags/anomaly-detection.md>), [attack-surface-management](<https://devfeed.tech/tags/attack-surface-management.md>), [authentication](<https://devfeed.tech/tags/authentication.md>), [automated](<https://devfeed.tech/tags/automated.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [behavioral-analytics](<https://devfeed.tech/tags/behavioral-analytics.md>), [blast-radius](<https://devfeed.tech/tags/blast-radius.md>), [exposure-management](<https://devfeed.tech/tags/exposure-management.md>), [external-attack-surface-management](<https://devfeed.tech/tags/external-attack-surface-management.md>), [incident-response](<https://devfeed.tech/tags/incident-response.md>), [infoblox](<https://devfeed.tech/tags/infoblox.md>), [kentik](<https://devfeed.tech/tags/kentik.md>), [network-observability](<https://devfeed.tech/tags/network-observability.md>), [network-traffic-analysis](<https://devfeed.tech/tags/network-traffic-analysis.md>), [patch](<https://devfeed.tech/tags/patch.md>), [php-vulnerability-management](<https://devfeed.tech/tags/php-vulnerability-management.md>), [php2shell](<https://devfeed.tech/tags/php2shell.md>), [platform](<https://devfeed.tech/tags/platform.md>), [protective-dns](<https://devfeed.tech/tags/protective-dns.md>), [security](<https://devfeed.tech/tags/security.md>), [systems](<https://devfeed.tech/tags/systems.md>), [teams](<https://devfeed.tech/tags/teams.md>), [vulnerability](<https://devfeed.tech/tags/vulnerability.md>), [vulnerability-management](<https://devfeed.tech/tags/vulnerability-management.md>)

### AI overview

PHP2Shell consists of two unauthenticated, remotely triggerable PHP core vulnerabilities that can enable code execution. The article urges patching fixed releases, upgrading unsupported PHP 8.1 and earlier, and finding exposed services; it says WAF rules are temporary mitigation, not a substitute. Infoblox describes complementary response capabilities: Exposure Management identifies internet-facing PHP, Universal Asset Insights can add asset and owner context, Threat Defense can block or detect malicious DNS activity after compromise, and Kentik helps investigate unusual external and internal traffic. The article explicitly says these tools cannot prevent the initial exploit payload from reaching a web server. Its reported EASM figures--fewer than 1% at fixed releases, about 62% on end-of-life branches and 38% on supported but vulnerable releases--are an exposure baseline, not an exploitation measure.

### Source excerpt

Last week's disclosure of PHP2Shell triggered an immediate, global race against the clock to address the vulnerability. As security teams scramble to patch affected systems, one thing is evident: enterprise patching is a marathon, not a sprint. The critical question isn't just how fast you can patch, but how you contain the blast radius today [...] The post PHP2Shell Is Already Here: How the Infoblox Platform Helps Uncover and Mitigate the Risk appeared first on Infoblox Blog.

## LABScon26 Replay | Keynote: A Sober Analysis of AI Impact

DevFeed: [LABScon26 Replay | Keynote: A Sober Analysis of AI Impact](<https://devfeed.tech/articles/labscon26-replay-keynote-a-sober-analysis-of-ai-impact-105726.md>)

Original publisher: [Read original article](<https://www.sentinelone.com/labs/labscon26-replay-keynote-a-sober-analysis-of-ai-impact/>)

Author: LABScon

Published: 2026-10-28T13:00:00Z

Content type: news

Language: en

Sources: [SentinelLabs](<https://devfeed.tech/sources/sentinellabs-we-are-hunters-reversers-exploit-developers-and-tinkerers-shedding-light-on-the-world-of-malware-exploits-apts-and-cybercrime-across-all-platforms.md>)

Topics: [Frontier AI](<https://devfeed.tech/topics/frontier-ai.md>), [Risk](<https://devfeed.tech/topics/risk.md>), [LABScon](<https://devfeed.tech/topics/labscon.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-impact](<https://devfeed.tech/tags/ai-impact.md>), [ai-investment](<https://devfeed.tech/tags/ai-investment.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [ai-research](<https://devfeed.tech/tags/ai-research.md>), [ai-risk](<https://devfeed.tech/tags/ai-risk.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [automation](<https://devfeed.tech/tags/automation.md>), [behaviour](<https://devfeed.tech/tags/behaviour.md>), [capex](<https://devfeed.tech/tags/capex.md>), [code](<https://devfeed.tech/tags/code.md>), [computer-science](<https://devfeed.tech/tags/computer-science.md>), [computer-security](<https://devfeed.tech/tags/computer-security.md>), [computers](<https://devfeed.tech/tags/computers.md>), [conference](<https://devfeed.tech/tags/conference.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [cpus](<https://devfeed.tech/tags/cpus.md>), [frontier-models](<https://devfeed.tech/tags/frontier-models.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [keynote](<https://devfeed.tech/tags/keynote.md>), [labscon](<https://devfeed.tech/tags/labscon.md>)

### AI overview

James Mickens argues that current AI investment may be disproportionate to its benefits, citing tech capital spending, rising model opacity, reported model behavior and sandbox escapes, and uncertain economic assumptions about jobs and universal basic income. He defines AI safety broadly to include economic and emotional security as well as computer security. His Guillotine hypervisor design runs model CPU code on isolated cores that cannot read or write the hardware state of the cores running the hypervisor. Mickens identifies hardware-based isolation of potentially misaligned frontier models as an important research direction for the next five years.

### Source excerpt

Harvard Professor James Mickens weighs the promises of AI against its potential harms, arguing for hardware-based mechanisms to isolate frontier models.

## What Is a Merchant Account? How It Works in 2026

DevFeed: [What Is a Merchant Account? How It Works in 2026](<https://devfeed.tech/articles/what-is-a-merchant-account-how-it-works-in-2026-104310.md>)

Original publisher: [Read original article](<https://dodopayments.com/blogs/what-is-a-merchant-account/>)

Author: Deepak Jangir

Published: 2026-10-28T00:00:00Z

Content type: article

Language: en

Sources: [Dodo Payments Blog](<https://devfeed.tech/sources/dodo-payments-blog.md>)

Topics: [Finance](<https://devfeed.tech/topics/finance.md>), [Risk](<https://devfeed.tech/topics/risk.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [about](<https://devfeed.tech/tags/about.md>), [acceptance](<https://devfeed.tech/tags/acceptance.md>), [account](<https://devfeed.tech/tags/account.md>), [added](<https://devfeed.tech/tags/added.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-tools](<https://devfeed.tech/tags/ai-tools.md>), [airlines](<https://devfeed.tech/tags/airlines.md>), [all](<https://devfeed.tech/tags/all.md>), [also](<https://devfeed.tech/tags/also.md>), [american-express](<https://devfeed.tech/tags/american-express.md>), [bank](<https://devfeed.tech/tags/bank.md>), [card](<https://devfeed.tech/tags/card.md>), [guide](<https://devfeed.tech/tags/guide.md>), [merchant-of-record](<https://devfeed.tech/tags/merchant-of-record.md>), [money](<https://devfeed.tech/tags/money.md>), [payments](<https://devfeed.tech/tags/payments.md>), [resources](<https://devfeed.tech/tags/resources.md>), [vs](<https://devfeed.tech/tags/vs.md>), [what-is](<https://devfeed.tech/tags/what-is.md>), [work](<https://devfeed.tech/tags/work.md>), [works](<https://devfeed.tech/tags/works.md>)

### AI overview

A merchant account is the acquiring-bank holding account where card proceeds arrive before payout to a business bank account; fees and chargebacks are deducted there, and the acquirer may hold reserves to cover risk. Businesses can use a dedicated account, an aggregator, or a merchant of record (MoR). Dedicated accounts involve upfront underwriting, contracts, and more integration work, but may offer negotiated pricing and greater control at high volume. Aggregators such as Stripe, PayPal, and Square offer fast setup and simple rates; businesses remain the legal merchant and may face payout pauses or reserves after later risk reviews. An MoR is the legal seller and handles tax compliance, fraud screening, and chargebacks, making it suited to digital businesses selling across countries. The article says the choice depends on processing volume, compliance capacity, and customer geography.

### Source excerpt

What is a merchant account? A plain-English guide to how merchant accounts work, how money flows from card to bank, and dedicated vs aggregator vs merchant of record.

## The web's having a rough week. Here's how to browse safely

DevFeed: [The web's having a rough week. Here's how to browse safely](<https://devfeed.tech/articles/the-web-s-having-a-rough-week-here-s-how-to-browse-safely-102403.md>)

Original publisher: [Read original article](<https://www.malwarebytes.com/blog/news/2026/10/the-webs-having-a-rough-week-heres-how-to-browse-safely>)

Author: Pieter Arntz

Published: 2026-10-27T12:22:37Z

Content type: article

Language: en

Sources: [Malwarebytes](<https://devfeed.tech/sources/malwarebytes.md>)

Topics: [Web](<https://devfeed.tech/topics/web.md>), [frontend security](<https://devfeed.tech/topics/frontend-security.md>), [Risk](<https://devfeed.tech/topics/risk.md>), [Google Chrome](<https://devfeed.tech/topics/google-chrome.md>)

Tags: [7-4](<https://devfeed.tech/tags/7-4.md>), [attacks](<https://devfeed.tech/tags/attacks.md>), [authentication](<https://devfeed.tech/tags/authentication.md>), [automatic](<https://devfeed.tech/tags/automatic.md>), [browser](<https://devfeed.tech/tags/browser.md>), [browser-extensions](<https://devfeed.tech/tags/browser-extensions.md>), [bugs](<https://devfeed.tech/tags/bugs.md>), [certificates](<https://devfeed.tech/tags/certificates.md>), [compromised-websites](<https://devfeed.tech/tags/compromised-websites.md>), [fake](<https://devfeed.tech/tags/fake.md>), [hacked](<https://devfeed.tech/tags/hacked.md>), [heres](<https://devfeed.tech/tags/heres.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [news](<https://devfeed.tech/tags/news.md>), [node-js](<https://devfeed.tech/tags/node-js.md>), [php](<https://devfeed.tech/tags/php.md>), [risk](<https://devfeed.tech/tags/risk.md>), [rogue](<https://devfeed.tech/tags/rogue.md>), [safely](<https://devfeed.tech/tags/safely.md>), [sites](<https://devfeed.tech/tags/sites.md>), [the-web](<https://devfeed.tech/tags/the-web.md>), [tls-certificates](<https://devfeed.tech/tags/tls-certificates.md>), [updates](<https://devfeed.tech/tags/updates.md>), [websites](<https://devfeed.tech/tags/websites.md>)

### AI overview

A familiar, genuine website can be compromised, so judge unexpected requests by whether they make sense for what you came to do. Don't run commands, install software, or change device settings just to pass a verification check; independently check update or account prompts through your browser's settings or an official app or support channel. Keep your operating system, browser, and apps updated, leave browser phishing and malware protection enabled, and use real-time anti-malware protection, unique passwords, and multifactor authentication. HTTPS protects the connection but does not establish that page content is safe. Browser warnings and security tools add protection, but cannot certify every unflagged page. The article cites attacks using hacked WordPress sites and fake verification pages, alongside separate PHP, Node.js Brotli, and certificate issues; it says most browsers blocked the compromised certificates, while other software and devices may still need updates.

### Source excerpt

PHP and Node.js flaws, hacked sites, fake updates, and rogue certificates can put you at risk even on trusted websites. Here's how to stay safe.

## PHP2Shell: How Kinsta protected WordPress sites

DevFeed: [PHP2Shell: How Kinsta protected WordPress sites](<https://devfeed.tech/articles/php2shell-how-kinsta-protected-wordpress-sites-98817.md>)

Original publisher: [Read original article](<https://kinsta.com/blog/php2shell-kinsta-wordpress/>)

Author: Joel Olawanle

Published: 2026-10-23T15:50:20Z

Content type: article

Language: en

Sources: [WordPress Guides, Tutorials & Tips - Kinsta® Blog](<https://devfeed.tech/sources/wordpress-guides-tutorials-tips-kinsta-blog.md>)

Topics: [WordPress](<https://devfeed.tech/topics/wordpress.md>), [Risk](<https://devfeed.tech/topics/risk.md>), [Memory Safety](<https://devfeed.tech/topics/memory-safety.md>), [CVSS v4.0](<https://devfeed.tech/topics/cvss-v4-0.md>)

Tags: [php2shell](<https://devfeed.tech/tags/php2shell.md>), [wordpress](<https://devfeed.tech/tags/wordpress.md>)

### AI overview

Kinsta says it protected hosted WordPress sites from PHP2Shell by enabling Cloudflare WAF protections and additional PHP configuration changes while preparing patches, then deploying security updates across PHP 7.4-8.5, including older-version fixes from Freexian. Because PHP2Shell affected the PHP runtime, WordPress, plugin, or theme updates alone would not fix it; Kinsta applied the patches at the hosting level, including to live, staging, and suspended environments. Kinsta reported no evidence of related breaches and marked the incident resolved on October 22, 2026. Its customers need take no PHP2Shell-specific action, though routine WordPress, plugin, and theme maintenance remains separate.

### Source excerpt

Featured image for PHP2Shell: How Kinsta protected WordPress sites

## Automotive Data Security: How 3 Brands Reduce Risk | Cyera

DevFeed: [Automotive Data Security: How 3 Brands Reduce Risk | Cyera](<https://devfeed.tech/articles/automotive-data-security-how-3-brands-reduce-risk-cyera-96447.md>)

Original publisher: [Read original article](<https://www.cyera.com/blog/leading-automotive-brands-shift-data-risk-management-into-high-gear-with-cyera>)

Author: Sam Dean

Published: 2026-10-22T22:39:01Z

Content type: article

Language: en

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

Topics: [Security](<https://devfeed.tech/topics/security.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>), [Risk](<https://devfeed.tech/topics/risk.md>), [data ownership](<https://devfeed.tech/topics/data-ownership.md>), [Access Control](<https://devfeed.tech/topics/access-control.md>)

Tags: [acquisitions](<https://devfeed.tech/tags/acquisitions.md>), [ai](<https://devfeed.tech/tags/ai.md>), [auction](<https://devfeed.tech/tags/auction.md>), [audits](<https://devfeed.tech/tags/audits.md>), [automotive](<https://devfeed.tech/tags/automotive.md>), [automotive-industry](<https://devfeed.tech/tags/automotive-industry.md>), [aws](<https://devfeed.tech/tags/aws.md>), [challenges](<https://devfeed.tech/tags/challenges.md>), [ciso](<https://devfeed.tech/tags/ciso.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-environments](<https://devfeed.tech/tags/cloud-environments.md>), [complex](<https://devfeed.tech/tags/complex.md>), [cyera](<https://devfeed.tech/tags/cyera.md>), [data](<https://devfeed.tech/tags/data.md>), [data-security](<https://devfeed.tech/tags/data-security.md>), [environments](<https://devfeed.tech/tags/environments.md>), [reduce](<https://devfeed.tech/tags/reduce.md>), [risk](<https://devfeed.tech/tags/risk.md>), [security](<https://devfeed.tech/tags/security.md>), [sensitive-data](<https://devfeed.tech/tags/sensitive-data.md>)

### AI overview

The excerpt presents Cyera as a way for automotive businesses to inventory sensitive data, identify exposure, and prioritize remediation across expanding environments. Valvoline reports a complete data inventory in 30 days, separation of retail and global operations data, and continuous monitoring after deployment in 10 minutes; it had been managing over 11 million unused records and 1,600 outdated accounts. At GPC, a proof of value found its sensitive-data estimates were off by 500%, uncovered $145 million in exposure in a subset of its estate, and recorded a 95% true positive rate. After deployment, GPC reports lowering risk by $45 million in six months, scanning all Office 365 in six weeks, and remediating 98% of cloud issues. ACV Auctions used Cyera to find cloud exposures including unencrypted PII and refine access policies; it says removing redundant data stores could save up to $500,000 annually. These are company-reported results in the excerpt.

### Source excerpt

See how Valvoline, GPC, and ACV Auctions use Cyera to discover sensitive data, reduce risk, and strengthen security across complex data environments.

## How Much Does One Wrong Answer Cost? Error Propagation in Constraint Pruning with Noisy LLM Oracles

DevFeed: [How Much Does One Wrong Answer Cost? Error Propagation in Constraint Pruning with Noisy LLM Oracles](<https://devfeed.tech/articles/how-much-does-one-wrong-answer-cost-error-propagation-in-constraint-pruning-with-noisy-llm-oracles-93540.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.25739>)

Author: Zhengtao Gui (University of California, Irvine), Weining Shen (University of California, Irvine)

Published: 2026-10-21T04:00:00Z

Content type: article

Language: en

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

Topics: [Error Propagation](<https://devfeed.tech/topics/error-propagation.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Risk](<https://devfeed.tech/topics/risk.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>)

Tags: [constraint](<https://devfeed.tech/tags/constraint.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [entity-resolution](<https://devfeed.tech/tags/entity-resolution.md>), [error](<https://devfeed.tech/tags/error.md>), [error-propagation](<https://devfeed.tech/tags/error-propagation.md>), [errors](<https://devfeed.tech/tags/errors.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [large-language-model](<https://devfeed.tech/tags/large-language-model.md>), [llm](<https://devfeed.tech/tags/llm.md>), [oracle](<https://devfeed.tech/tags/oracle.md>), [oracles](<https://devfeed.tech/tags/oracles.md>), [queries](<https://devfeed.tech/tags/queries.md>), [query](<https://devfeed.tech/tags/query.md>), [random](<https://devfeed.tech/tags/random.md>), [records](<https://devfeed.tech/tags/records.md>), [risk](<https://devfeed.tech/tags/risk.md>), [rules](<https://devfeed.tech/tags/rules.md>), [sampling](<https://devfeed.tech/tags/sampling.md>), [simulations](<https://devfeed.tech/tags/simulations.md>), [taxonomies](<https://devfeed.tech/tags/taxonomies.md>)

### AI overview

Constraint pruning can spread an accepted wrong oracle answer to every fact it implies, so its first-order risk depends on the average number of facts corrupted per error. For implication rules under random query order, the effects balance: an error that would change k facts is accepted with probability 1/k, costing one fact on average. Rules with s premises can break this balance; an error at a premise shared by r rules can cost on the order of r^(1−1/s) facts. For entity resolution among m records of one entity, first-order risk is ε times the Wiener index of the spanning tree induced by query order. Pivot orders are optimal and amplify errors by less than 2; random orders amplify them by order m^(1/3). Most-similar-first is worst in one dimension and, for large m, worse than random in two. With multiple entities, the cost of a wrong merge does not depend on how many others exist. Simulations match the predicted constants within sampling error.

### Source excerpt

arXiv:2610.25739v1 Announce Type: new Abstract: Constraint pruning saves queries to a large language model (LLM) or another noisy oracle by skipping each yes/no question implied by answers accepted earlier, for example through transitivity. Entity-resolution systems and LLM-based rankers use this idea. However, their analyses typically assume a correct oracle. We argue that pruning should be judged by its risk when the oracle errs, since an accepted wrong answer spreads to every question it implies. When each answer is wrong independently with small probability $\varepsilon$, the first-order risk is set by how many facts one wrong answer corrupts on average. We consider two kinds of constraints. For implication rules, as in taxonomies, pruning in a random order adds no first-order risk: a wrong answer that would change $k$ facts is accepted with probability $1/k$, so it costs one fact on average. Rules with $s$ premises break this balance: a wrong answer at a premise shared by $r$ rules can cost order $r^{1-1/s}$ facts. For entity resolution with $m$ records of one entity, the first-order risk equals $\varepsilon$ times the Wiener index (the total pairwise distance) of the spanning tree built by the query order. Pivot orders, which first compare one record with all others, amplify errors by less than 2 and are optimal; random orders by order $m^{1/3}$; the common most-similar-first order is the worst in one dimension and, for large $m$, worse than random in two. With several entities, the cost of wrongly merging two does not depend on how many others exist. Simulations agree with every predicted constant within sampling error.

## A Digital Twin to Improve Cyber-Social Impact

DevFeed: [A Digital Twin to Improve Cyber-Social Impact](<https://devfeed.tech/articles/a-digital-twin-to-improve-cyber-social-impact-93839.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.28430>)

Author: Vita Santa Barletta, Danilo Caivano, Gabriel Cellammare, Andrea Dimauro, Pasquale Mesagna, Massimiliano Morga

Published: 2026-10-21T04:00:00Z

Content type: article

Language: en

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

Topics: [AI in cybersecurity](<https://devfeed.tech/topics/ai-in-cybersecurity.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Risk](<https://devfeed.tech/topics/risk.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [ai](<https://devfeed.tech/tags/ai.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [analyst](<https://devfeed.tech/tags/analyst.md>), [analyze](<https://devfeed.tech/tags/analyze.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [assessment](<https://devfeed.tech/tags/assessment.md>), [at-the-core](<https://devfeed.tech/tags/at-the-core.md>), [attack](<https://devfeed.tech/tags/attack.md>), [automate](<https://devfeed.tech/tags/automate.md>), [availability](<https://devfeed.tech/tags/availability.md>), [based](<https://devfeed.tech/tags/based.md>), [but](<https://devfeed.tech/tags/but.md>), [capture](<https://devfeed.tech/tags/capture.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cyber](<https://devfeed.tech/tags/cyber.md>), [digital](<https://devfeed.tech/tags/digital.md>), [digital-twin](<https://devfeed.tech/tags/digital-twin.md>), [impact](<https://devfeed.tech/tags/impact.md>), [social](<https://devfeed.tech/tags/social.md>), [social-impact](<https://devfeed.tech/tags/social-impact.md>)

### AI overview

The proposed framework combines Digital Twins with a Cyber-Social Security model to assess cyber threats in critical infrastructure across technical and social dimensions. It uses virtualized ICS environments built with Mininet and MiniCPS to model attacks, and introduces CSVSS, which extends CVSS with a Social Impact dimension based on the Entity and Place pillars to produce a reproducible severity score. An LLM performs first-pass vulnerability mapping and drafts risk reports; a human analyst reviews the findings with explainable AI feedback and retains decision authority. The reported experiments demonstrate cyber-social impact modeling and scoring, while human oversight is presented as a conceptual design principle.

### Source excerpt

arXiv:2610.28430v1 Announce Type: new Abstract: The increasing complexity of Industrial Control Systems (ICS) and their integration with IT infrastructures has significantly expanded the cyber threat landscape. Traditional cybersecurity models, primarily based on the Confidentiality, Integrity, and Availability (CIA triad), fail to capture the broader societal and operational impacts of cyberattacks. This paper proposes a novel framework that integrates Digital Twin technology with a Cyber-Social Security model to simulate, analyze, and quantify cyber threats in critical infrastructures. Virtualized ICS environments built with Mininet and MiniCPS are used to model attack scenarios and assess their consequences across technical and social dimensions. At the core of the framework lies the Cyber-Social Vulnerability Scoring System (CSVSS), a quantitative metric that extends CVSS with a Social Impact dimension grounded in the Entity and Place pillars, producing a single, reproducible severity score that reflects real-world consequences. Crucially, the framework does not automate risk assessment but keeps the human analyst at its center: a Large Language Model (LLM) performs a first-pass analysis, mapping vulnerabilities and drafting risk reports, while the analyst, supported by explainable AI feedback, validates the findings and retains decision authority over critical cases. Experimental results demonstrate the framework's ability to model and score cyber-social impact, while the human-in-the-loop design is presented as a conceptual principle to preserve oversight.

## Drupal Warns of Fast Exploitation Risk for Critical Core RCE Flaw

DevFeed: [Drupal Warns of Fast Exploitation Risk for Critical Core RCE Flaw](<https://devfeed.tech/articles/drupal-warns-of-fast-exploitation-risk-for-critical-core-rce-flaw-90049.md>)

Original publisher: [Read original article](<https://www.securityweek.com/drupal-warns-of-fast-exploitation-risk-for-critical-core-rce-flaw/>)

Author: Ionut Arghire

Published: 2026-10-19T15:47:50Z

Content type: news

Language: en

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

Topics: [Drupal](<https://devfeed.tech/topics/drupal.md>), [known exploitable vulnerabilities](<https://devfeed.tech/topics/known-exploitable-vulnerabilities.md>), [Risk](<https://devfeed.tech/topics/risk.md>), [bug](<https://devfeed.tech/topics/bug.md>)

Tags: [critical](<https://devfeed.tech/tags/critical.md>), [drupal](<https://devfeed.tech/tags/drupal.md>), [rce](<https://devfeed.tech/tags/rce.md>), [remote-code-execution](<https://devfeed.tech/tags/remote-code-execution.md>), [security](<https://devfeed.tech/tags/security.md>), [security-updates](<https://devfeed.tech/tags/security-updates.md>), [vulnerabilities](<https://devfeed.tech/tags/vulnerabilities.md>), [vulnerability](<https://devfeed.tech/tags/vulnerability.md>)

### AI overview

Drupal released fixes for five vulnerabilities, including two critical flaws in the Entity List and core Field UI modules with exploit paths involving contributed modules. The article also reports cross-site scripting fixes and advises users to update to patched Drupal and module versions. Drupal said it had no evidence of in-the-wild exploitation of these flaws.

### Source excerpt

Critical flaws impact the Entity List and Field UI modules, and exploit paths exist with contributed modules. The post Drupal Warns of Fast Exploitation Risk for Critical Core RCE Flaw appeared first on SecurityWeek.

## Partial Refund: Meaning, Examples, and How to Issue One

DevFeed: [Partial Refund: Meaning, Examples, and How to Issue One](<https://devfeed.tech/articles/partial-refund-meaning-examples-and-how-to-issue-one-88957.md>)

Original publisher: [Read original article](<https://dodopayments.com/blogs/partial-refund/>)

Author: Deepak Jangir

Published: 2026-10-17T00:00:00Z

Content type: article

Language: en

Sources: [Dodo Payments Blog](<https://devfeed.tech/sources/dodo-payments-blog.md>)

Topics: [Finance](<https://devfeed.tech/topics/finance.md>), [Use Cases for Every Cloud Service Model](<https://devfeed.tech/topics/use-cases-for-every-cloud-service-model.md>), [Service-Level Agreements](<https://devfeed.tech/topics/service-level-agreements.md>), [Risk](<https://devfeed.tech/topics/risk.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-products](<https://devfeed.tech/tags/ai-products.md>), [api](<https://devfeed.tech/tags/api.md>), [b2b](<https://devfeed.tech/tags/b2b.md>), [chargeback](<https://devfeed.tech/tags/chargeback.md>), [chargebacks](<https://devfeed.tech/tags/chargebacks.md>), [digital](<https://devfeed.tech/tags/digital.md>), [examples](<https://devfeed.tech/tags/examples.md>), [fees](<https://devfeed.tech/tags/fees.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [issue](<https://devfeed.tech/tags/issue.md>), [meaning](<https://devfeed.tech/tags/meaning.md>), [payments](<https://devfeed.tech/tags/payments.md>), [refunds](<https://devfeed.tech/tags/refunds.md>), [saas](<https://devfeed.tech/tags/saas.md>), [sales](<https://devfeed.tech/tags/sales.md>), [tax](<https://devfeed.tech/tags/tax.md>), [vat](<https://devfeed.tech/tags/vat.md>)

### AI overview

A partial refund returns a specific portion of a settled payment when a customer received some, but not all, of the value; it differs from a void, chargeback, account credit, and discount. The article gives examples including prorated downgrades, affected items in a bundle, outages, and goodwill gestures, and recommends full refunds for cases such as duplicate charges or a product that never worked. It says tax should generally be refunded proportionally. For Dodo Payments, the original processing fee is not reversed, each refund costs $1, refunds require a successful payment within the default 30-day window and sufficient wallet balance, and only one can be in progress per payment. Refunds can be issued per item in the dashboard or via `POST /refunds`; the API amount uses the smallest currency unit. The article also advises waiting for `refund.succeeded` or `refund.failed` webhooks before changing app access or account state, and notes a partial refund may not prevent a dispute if the customer expected a full refund.

### Source excerpt

What a partial refund is, when SaaS and digital sellers should use one, how fees and sales tax or VAT are handled, and how to issue a partial refund step by step.

## How to Retrieve a Phone Number's SMS Pumping Score in Rust

DevFeed: [How to Retrieve a Phone Number's SMS Pumping Score in Rust](<https://devfeed.tech/articles/how-to-retrieve-a-phone-number-s-sms-pumping-score-in-rust-86803.md>)

Original publisher: [Read original article](<https://www.twilio.com/en-us/blog/developers/tutorials/retrieve-phone-number-sms-pumping-risk-score-rust>)

Author: Matthew Setter

Published: 2026-10-14T00:00:00Z

Content type: tutorial

Language: en

Sources: [Twilio](<https://devfeed.tech/sources/twilio-blog.md>), [Twilio Blog](<https://devfeed.tech/sources/twilio-blog-2.md>)

Topics: [Rust](<https://devfeed.tech/topics/rust.md>), [API](<https://devfeed.tech/topics/api.md>), [Risk](<https://devfeed.tech/topics/risk.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [developer-insights](<https://devfeed.tech/tags/developer-insights.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [lookup](<https://devfeed.tech/tags/lookup.md>), [pumping-risk-score](<https://devfeed.tech/tags/pumping-risk-score.md>), [risk](<https://devfeed.tech/tags/risk.md>), [rust](<https://devfeed.tech/tags/rust.md>)

### AI overview

This tutorial shows how to use Rust and Twilio's Lookup v2 API to retrieve a phone number's SMS Pumping Risk Score and related information. It describes the fraud risk the score is intended to help assess and outlines a project setup and code workflow.

### Source excerpt

In this tutorial, you'll learn how to retrieve a phone number's SMS Pumping Risk Score -- along with other helpful information -- in Rust.

## Responsible AI governance: How AWS positions customers to align with ISO/IEC 42005:2025

DevFeed: [Responsible AI governance: How AWS positions customers to align with ISO/IEC 42005:2025](<https://devfeed.tech/articles/responsible-ai-governance-how-aws-positions-customers-to-align-with-iso-iec-42005-2025-65900.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/machine-learning/responsible-ai-governance-how-aws-positions-customers-to-align-with-iso-iec-420052025/>)

Author: Adam Powers

Published: 2026-10-06T15:53:28Z

Content type: article

Language: en

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

Topics: [ai-governance](<https://devfeed.tech/topics/ai-governance.md>), [responsible-ai](<https://devfeed.tech/topics/responsible-ai.md>), [Responsibility & Safety](<https://devfeed.tech/topics/responsibility-safety.md>), [Risk](<https://devfeed.tech/topics/risk.md>), [Web](<https://devfeed.tech/topics/web.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-governance](<https://devfeed.tech/tags/ai-governance.md>), [governance](<https://devfeed.tech/tags/governance.md>), [intermediate-200](<https://devfeed.tech/tags/intermediate-200.md>), [responsible-ai](<https://devfeed.tech/tags/responsible-ai.md>), [risk](<https://devfeed.tech/tags/risk.md>), [standards](<https://devfeed.tech/tags/standards.md>), [thought-leadership](<https://devfeed.tech/tags/thought-leadership.md>)

### AI overview

This AWS post explains how ISO/IEC 42005:2025 guides organizations in scoping, conducting, documenting, and reviewing AI system impact assessments. It describes how assessments can fit into enterprise risk processes, identify potential benefits and harms, and inform risk treatment throughout the AI development lifecycle. AWS also describes its Responsible AI Lens as a tool aligned with the standard.

### Source excerpt

AWS invests in tools that help customers align with international standards for responsible AI governance. In this post, we explore the AI system impact assessment: what it is, how it improves enterprise-wide risk management, and how ISO/IEC 42005:2025 codifies best practices for conducting and documenting these assessments.

## How to Reduce Overprivileged Kubernetes Service Accounts

DevFeed: [How to Reduce Overprivileged Kubernetes Service Accounts](<https://devfeed.tech/articles/how-to-reduce-overprivileged-kubernetes-service-accounts-65859.md>)

Original publisher: [Read original article](<https://goteleport.com/blog/how-to-reduce-overprivileged-kubernetes-service-accounts/>)

Author: info@goteleport.com (Justin O'Leary)

Published: 2026-10-06T00:00:00Z

Content type: tutorial

Language: en

Sources: [Teleport](<https://devfeed.tech/sources/teleport.md>)

Topics: [Risk](<https://devfeed.tech/topics/risk.md>), [Credential theft](<https://devfeed.tech/topics/credential-theft.md>), [Helm charts](<https://devfeed.tech/topics/helm-charts.md>)

Tags: [authentication](<https://devfeed.tech/tags/authentication.md>), [authorization](<https://devfeed.tech/tags/authorization.md>), [automation](<https://devfeed.tech/tags/automation.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [kubernetes](<https://devfeed.tech/tags/kubernetes.md>), [service-accounts](<https://devfeed.tech/tags/service-accounts.md>)

### AI overview

The tutorial explains how broad RBAC bindings, cloud IAM permissions, shared automation, and long-lived credentials sustain excessive Kubernetes ServiceAccount access. It provides commands and configuration examples for auditing privileges, disabling unnecessary token mounting, replacing static credentials, and restricting workloads to required permissions. It also recommends reviewing Kubernetes and cloud permissions together and automating RBAC consistency across clusters.

### Source excerpt

Learn how to reduce Kubernetes ServiceAccount risk.

## AVS Mismatch: What It Means and How to Reduce Declines

DevFeed: [AVS Mismatch: What It Means and How to Reduce Declines](<https://devfeed.tech/articles/avs-mismatch-what-it-means-and-how-to-reduce-declines-63692.md>)

Original publisher: [Read original article](<https://dodopayments.com/blogs/avs-mismatch/>)

Author: Ayush Agarwal

Published: 2026-10-02T00:00:00Z

Content type: article

Language: en

Sources: [Dodo Payments Blog](<https://devfeed.tech/sources/dodo-payments-blog.md>)

Topics: [Finance](<https://devfeed.tech/topics/finance.md>), [Risk](<https://devfeed.tech/topics/risk.md>)

Tags: [attack](<https://devfeed.tech/tags/attack.md>), [austin](<https://devfeed.tech/tags/austin.md>), [bank](<https://devfeed.tech/tags/bank.md>), [banking](<https://devfeed.tech/tags/banking.md>), [billing](<https://devfeed.tech/tags/billing.md>), [canada](<https://devfeed.tech/tags/canada.md>), [card](<https://devfeed.tech/tags/card.md>), [chargeback](<https://devfeed.tech/tags/chargeback.md>), [chargebacks](<https://devfeed.tech/tags/chargebacks.md>), [checkout](<https://devfeed.tech/tags/checkout.md>), [churn](<https://devfeed.tech/tags/churn.md>), [code](<https://devfeed.tech/tags/code.md>), [customer](<https://devfeed.tech/tags/customer.md>), [dashboard](<https://devfeed.tech/tags/dashboard.md>), [days](<https://devfeed.tech/tags/days.md>), [design](<https://devfeed.tech/tags/design.md>), [documentation](<https://devfeed.tech/tags/documentation.md>), [dodo](<https://devfeed.tech/tags/dodo.md>), [email](<https://devfeed.tech/tags/email.md>), [fraud-prevention](<https://devfeed.tech/tags/fraud-prevention.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [payments](<https://devfeed.tech/tags/payments.md>)

### AI overview

AVS compares a card's billing street number and postal code with issuer records, then returns a result; the issuer or merchant decides whether to approve or decline. A mismatch can reflect an honest typo, move, address-format issue, or limited international support, so it is a noisy fraud signal, not proof of fraud. The article recommends improving billing-address fields, accepting partial matches and unavailable results unless other risk signals arise, treating no-match results as higher risk, and using CVV, behavioral checks, or 3D Secure to step up risky orders. AVS does not shift chargeback liability; successful 3DS generally can. For Dodo Payments, the article advises routing payment failures by documented error codes and distinguishing hard from soft declines rather than guessing an AVS code.

### Source excerpt

An AVS mismatch means the billing address entered at checkout did not match the card issuer's records. Learn AVS codes, why it declines, and how to fix it.

## The Business Case for AI Code Review: Costs, ROI, and How to Measure Impact

DevFeed: [The Business Case for AI Code Review: Costs, ROI, and How to Measure Impact](<https://devfeed.tech/articles/the-business-case-for-ai-code-review-costs-roi-and-how-to-measure-impact-61148.md>)

Original publisher: [Read original article](<https://hackernoon.com/the-business-case-for-ai-code-review-costs-roi-and-how-to-measure-impact?source=rss>)

Author: Qodo

Published: 2026-09-28T12:13:31Z

Content type: article

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [ai code review](<https://devfeed.tech/topics/ai-code-review.md>), [developer-productivity](<https://devfeed.tech/topics/developer-productivity.md>), [Risk](<https://devfeed.tech/topics/risk.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-code-review](<https://devfeed.tech/tags/ai-code-review.md>), [ai-code-review-platform](<https://devfeed.tech/tags/ai-code-review-platform.md>), [ai-code-review-roi](<https://devfeed.tech/tags/ai-code-review-roi.md>), [ai-coding-tools-business-case](<https://devfeed.tech/tags/ai-coding-tools-business-case.md>), [ai-generated-code](<https://devfeed.tech/tags/ai-generated-code.md>), [ai-generated-code-quality](<https://devfeed.tech/tags/ai-generated-code-quality.md>), [ai-software-quality-assurance](<https://devfeed.tech/tags/ai-software-quality-assurance.md>), [code-review](<https://devfeed.tech/tags/code-review.md>), [costs](<https://devfeed.tech/tags/costs.md>), [engineering-productivity](<https://devfeed.tech/tags/engineering-productivity.md>), [good-company](<https://devfeed.tech/tags/good-company.md>), [pull-request-automation](<https://devfeed.tech/tags/pull-request-automation.md>), [roi](<https://devfeed.tech/tags/roi.md>), [software-engineering](<https://devfeed.tech/tags/software-engineering.md>), [tests](<https://devfeed.tech/tags/tests.md>)

### AI overview

AI coding tools increase output, but bugs, rework, and production incidents can shift costs downstream. The article explains how engineering leaders can build a business case for AI code review and measure its impact.

### Source excerpt

AI coding tools boost developer output, but hidden costs appear later through bugs, rework, and production incidents. Learn how AI code review protects ROI.

## Reduce Manual Review While Improving Approval & Acceptance Rates

DevFeed: [Reduce Manual Review While Improving Approval & Acceptance Rates](<https://devfeed.tech/articles/reduce-manual-review-while-improving-approval-acceptance-rates-60171.md>)

Original publisher: [Read original article](<https://sift.com/blog/reduce-manual-review/>)

Author: Ben Price

Published: 2026-09-25T20:00:00Z

Content type: article

Language: en

Sources: [Sift Science](<https://devfeed.tech/sources/sift-science.md>)

Topics: [trust and safety](<https://devfeed.tech/topics/trust-and-safety.md>), [Security, Privacy and Abuse Prevention](<https://devfeed.tech/topics/security-privacy-and-abuse-prevention.md>), [Risk](<https://devfeed.tech/topics/risk.md>), [data](<https://devfeed.tech/topics/data.md>)

Tags: [acceptance](<https://devfeed.tech/tags/acceptance.md>), [acceptance-rate](<https://devfeed.tech/tags/acceptance-rate.md>), [approval-rate](<https://devfeed.tech/tags/approval-rate.md>), [customer-experience](<https://devfeed.tech/tags/customer-experience.md>), [digital-trust](<https://devfeed.tech/tags/digital-trust.md>), [dynamic-friction](<https://devfeed.tech/tags/dynamic-friction.md>), [fraud](<https://devfeed.tech/tags/fraud.md>), [fraud-decisioning](<https://devfeed.tech/tags/fraud-decisioning.md>), [fraud-programs](<https://devfeed.tech/tags/fraud-programs.md>), [manual-review](<https://devfeed.tech/tags/manual-review.md>), [queues](<https://devfeed.tech/tags/queues.md>), [safety](<https://devfeed.tech/tags/safety.md>), [trust-and-safety](<https://devfeed.tech/tags/trust-and-safety.md>), [trust-safety](<https://devfeed.tech/tags/trust-safety.md>)

### AI overview

This article examines how growing manual fraud-review queues increase labor costs, slow checkout and account approval, and contribute to false declines. It argues that mature fraud programs should improve decisioning rather than simply expand review volume.

### Source excerpt

Over time, Trust and Safety teams deal with a problem where the more fraud that's caught, the more manual review queues grow. Technically, they can remedy this issue by cutting review volume to protect the customer experience, but when they do this, false declines will creep up instead. Recent industry data suggests this doesn't need [...] The post Reduce Manual Review While Improving Approval & Acceptance Rates appeared first on Sift.

[Next page](<https://devfeed.tech/topics/risk.md?cursor=WyIyMDI2LTA5LTI2VDAxOjIwOjE1LjQxMDYwOSswMDowMCIsICI1ZTMxN2IxNi01YjYyLTQ5MmItYjJmZi1hODNjOTQ0MzVmMDYiXQ%3D%3D>)