# facebook Published articles for facebook. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## Meta does a U-turn on controversial new AI terms for businesses DevFeed: [Meta does a U-turn on controversial new AI terms for businesses]() Original publisher: [Read original article]() Author: Maxwell Cooter Published: 2026-11-06T14:54:42Z Content type: news Language: en Sources: [Computerworld]() Topics: [X (Twitter)](), [AI Chat]() Tags: [ai](), [ai-terms](), [artificial-intelligence](), [business](), [content](), [customers](), [data](), [delay](), [edition-us](), [employees](), [error](), [facebook](), [harvard](), [illegal](), [language-en](), [legal](), [meta](), [news-brief](), [note](), [october](), [regulators](), [statement](), [storytype-news](), [terms](), [uk]() ### AI overview Meta indefinitely paused proposed terms for business customers after pushback. The terms would have allowed Meta to take content from customers' websites and made customers responsible for errors caused by faulty AI. Announced October 19, the changes were first scheduled for October 30 and later delayed to November 20; Meta then said it would pause them while engaging with stakeholders. The article says customers viewed the terms as onerous and potentially exposing them to legal risks. ### Source excerpt Meta has hit pause on new terms of service for its business customers that would have granted it the right to grab any content from their websites. Not only that, but it would also have shifted the blame for every error produced through faulty AI to its customers. The new terms of service were revealed on October 19 and initially due to take effect on October 30, though Meta later pushed the start date back to November 20. But that small delay wasn't enough to calm customer pushback, and now Meta has posted a note on Facebook's business help center saying that the update had now been paused indefinitely. Given the strength of feeling, Meta's U-turn will not have come as a surprise. The terms would have been seen as far too onerous for customers, as well as exposing them to a legal minefield of the sort that UK regulators recently warned about. In a statement to Harvard's Nieman Lab, the company said that "We're pausing the changes as we continue to engage with stakeholders and take our time to get this right." Meta has had a troubled time recently. Two weeks ago, it was fined for illegal use of customers' data and in May, the company laid off 8,000 employees. This article first appeared on CIO. ## Everyone Everywhere Hated Meta's New AI Training Terms. Now, They're On Pause DevFeed: [Everyone Everywhere Hated Meta's New AI Training Terms. Now, They're On Pause]() Original publisher: [Read original article]() Author: Samantha Cole Published: 2026-11-05T21:01:55Z Content type: news Language: en Sources: [404 Media]() Topics: [AI-Ready Data](), [post-training]() Tags: [advertising](), [ai](), [ai-products](), [ai-terms](), [ai-training](), [business](), [competition-and-markets-authority](), [content](), [court](), [data](), [domains](), [facebook](), [generative-ai](), [injunction](), [instagram](), [investigation](), [media](), [meta](), [terms](), [training]() ### AI overview Meta has paused planned commercial-term changes that would have let it use content from business accounts' domains, web pages, or URLs to develop and improve products, among other purposes. The changes will not take effect on November 20, 2026; existing terms remain in effect while Meta consults stakeholders. The excerpt reports that a Norwegian court temporarily enjoined the terms and the UK Competition and Markets Authority opened an investigation. It also says Meta had earlier delayed the change and narrowed the stated scope to advertising and business products, removing "train" from the listed uses. ### Source excerpt Back to the drawing board! ## Backup codes: How to store them so you're never locked out DevFeed: [Backup codes: How to store them so you're never locked out]() Original publisher: [Read original article]() Author: Jessica Bernard Published: 2026-11-03T15:22:59Z Content type: tutorial Language: en Sources: [The Proton Blog]() Topics: [password reset](), [backups](), [information-technology]() Tags: [2fa](), [account](), [account-security](), [authentication](), [backup](), [backup-verification](), [best-practices](), [business](), [cloud](), [cloud-storage](), [code](), [email](), [encrypted](), [encryption](), [facebook](), [for-business](), [generate](), [google](), [how-to](), [recovery](), [store]() ### AI overview Backup codes provide single-use access when a second factor is unavailable, but anyone who obtains one may bypass that factor with the account password. Store copies in separate secure locations: use an encrypted password manager for digital storage and a locked physical location for a printout. Avoid plain notes, email drafts, screenshots, and storing codes on the device or service they protect. Generate a replacement set in account security settings if codes are lost or exposed; doing so usually invalidates the old set. Codes generally do not expire on their own, but each works only once. Never share one with someone claiming to be support; enter it only on the sign-in screen. ### Source excerpt Lock out of your account? Backup codes are your way back in. Learn where recovery codes live, how to store them, and how to generate new ones. ## Google and Meta Open-source TorchTPU to Bring Native PyTorch Support to Google TPUs DevFeed: [Google and Meta Open-source TorchTPU to Bring Native PyTorch Support to Google TPUs]() Original publisher: [Read original article]() Author: Sergio De Simone Published: 2026-11-02T17:00:00Z Content type: news Language: en Sources: [InfoQ]() Topics: [PyTorch](), [distributed-training]() Tags: [agents](), [ai](), [ai-ml-data-engineering](), [development](), [facebook](), [google](), [gpu](), [large-language-models](), [meta](), [ml-data-engineering](), [news](), [open-source](), [pytorch](), [torchtpu](), [torchtpu-native-pythorch-tpu](), [tpus]() ### AI overview TorchTPU, an open-source Google and Meta project, brings native PyTorch execution to Google TPUs with minor migration changes. Compared with PyTorch/XLA, which the article says required explicit compilable regions, had high compile latency, lacked native eager execution, and lagged in distributed primitives, TorchTPU prioritizes eager execution and also supports compiled and distributed workloads. GPU migrations can replace the `cuda` device with `tpu` and `nccl` with `tpu_dist`, or use `torch.accelerator.current_accelerator`. TorchTPU offers strict, debug, and DeferAndFuse eager modes; Google claims DeferAndFuse can be at least 50% faster than Strict Eager while reducing peak HBM memory. Google evaluated about 5,000 Hugging Face models across 500+ architectures, and reported Muse Glimmer throughput and large-scale Qwen3 MoE training results on TPUv7. The project remains early-stage, with dynamic shapes, compilation performance, operator coverage, reinforcement learning, and quantization on its roadmap. ### Source excerpt TorchTPU is an open-source project developed by Google in collaboration with Meta that provides a native PyTorch backend for Google TPUs. It enables PyTorch workloads to execute on TPU hardware with minimal code modifications, supporting eager and compiled execution, distributed computing, and hardware-specific optimizations. By Sergio De Simone ## I'm Finally AI-Pilled DevFeed: [I'm Finally AI-Pilled]() Original publisher: [Read original article]() Author: Kevin Naughton Jr. Published: 2026-10-29T15:15:56Z Content type: opinion Language: en Sources: [The Software Engineer Weekly]() Topics: [Artificial Intelligence](), [AI Chat]() Tags: [acceptance](), [ads](), [ai](), [better](), [coding](), [email](), [engineering](), [facebook](), [facebook-marketplace](), [funnel](), [goals](), [google](), [history](), [identity](), [information](), [instagram](), [job](), [knowledge](), [learning](), [leverage]() ### AI overview The author argues that accepting AI's strengths beyond one's own expertise can shift focus from protecting a job identity to reaching personal goals faster. Although they still claim an edge in some coding and architecture decisions, they see AI's knowledge advantage in other fields as growing. As an example, the author used AI through several back-and-forth discussions to create a marketing funnel and email sequence in one night, then launched ads for Ferryman.io on TikTok, Instagram, Facebook, and Google despite having little marketing experience. They present this as an experiment, not a proven successful campaign. ### Source excerpt It's taken a long time, but I'm finally AI-pilled. I've finally realized that there are simply things that AI is much better at doing than me. Lots of people don't like to admit this but it's becoming more and more true each day ## Nvidia, AMD to buy $4B in Meta shares as part of expanded chip partnerships DevFeed: [Nvidia, AMD to buy $4B in Meta shares as part of expanded chip partnerships]() Original publisher: [Read original article]() Author: Maria Deutscher Published: 2026-10-27T21:12:55Z Content type: news Language: en Sources: [SiliconANGLE]() Topics: [Nvidia](), [NVIDIA Vera Rubin](), [NVIDIA Spectrum-X Ethernet](), [Chip design](), [Frontier AI](), [Spectrum Ethernet](), [Image]() Tags: [advanced](), [advanced-micro-devices](), [also](), [amd](), [chip](), [chipmakers](), [cpu](), [data-center](), [devices](), [equity](), [facebook](), [fundraising](), [future](), [gpu](), [inc](), [infra](), [initiative](), [kick](), [meta](), [meta-platforms](), [nasdaq-meta](), [news](), [nvidia](), [nvidia-corp](), [partnerships](), [rubin](), [shares](), [spectrum-x-ethernet](), [the-latest](), [top-story-4](), [top-story-5](), [vera]() ### AI overview Meta says Nvidia and AMD will buy $4 billion in its shares as part of a planned stock sale of up to $95 billion; together with Broadcom, the chipmakers plan to invest $5 billion, with Nvidia accounting for $3 billion and AMD $1 billion. The expanded Nvidia partnership includes wider use of Rubin GPUs, Vera CPUs and Spectrum-X networking, work to improve Meta's Muse models and AI agent, and exploration of Nvidia's OpenShell agent sandbox architecture. AMD says Meta has begun testing initial Helios racks built around Instinct MI450 GPUs, Epyc CPUs and Pensando data processing units, with further deployments planned. Meta also expects to adopt a custom AMD accelerator later; AMD says partnership data will inform future chip and rack designs. ### Source excerpt Meta Platforms Inc. today announced that Nvidia Corp. and Advanced Micro Devices Inc. plan to buy $4 billion worth of its shares. The chipmakers will purchase the equity as part of an up to $95 billion stock sale that the Facebook parent intends to kick off in the near future. The fundraising initiative will also [...] The post Nvidia, AMD to buy $4B in Meta shares as part of expanded chip partnerships appeared first on SiliconANGLE. ## Silicon Valley parents want the tech they help build out of their children's schools DevFeed: [Silicon Valley parents want the tech they help build out of their children's schools]() Original publisher: [Read original article]() Author: Robin Buller in San Francisco Published: 2026-10-27T12:00:38Z Content type: news Language: en Sources: [The Guardian | Technology]() Topics: [Learning]() Tags: [2015](), [area](), [build](), [but](), [california](), [california-silicon-valley-us-education-education-children-children-s-tech-society-technology-us-news](), [children](), [children-s-tech](), [education](), [engineer](), [experience](), [facebook](), [gil](), [github](), [help](), [moving](), [netflix](), [one](), [private](), [silicon](), [silicon-valley](), [society](), [tech](), [technology](), [us-education](), [us-news](), [want](), [west-coast]() ### AI overview Tech industry parents in affluent California communities are pressing schools to reduce classroom screen use, citing concerns about learning, student-teacher relationships and young children's exposure. San Mateo schools stopped sending Chromebooks home with students in kindergarten through second grade and banned high schoolers' cellphones; Mountain View is moving away from i-Ready, while Palo Alto prohibited smart glasses. Parents also want districts to vet educational products and require evidence of learning outcomes. Private schools such as San Francisco Waldorf can change policies more quickly and devote more resources to teachers, but its tuition is high. Educators warn that lower-income districts may rely more heavily on screens when budgets are tight, while families leaving public schools for private options can take funding with them. The excerpt describes local efforts and concerns, not their long-term effects. ### Source excerpt Both private and public schools in tech regions are limiting the use of tech in classrooms. But will low-income districts be able to follow suit? German Gil spends most of his waking hours looking at screens. Since moving to the San Francisco Bay Area in 2015, the software engineer has bounced between startups and giants including Netflix, Facebook and GitHub. But when it came to the education of his daughter, Victoria, he wanted her to have a different experience, one in which his "kid could be a kid". Continue reading... ## AI-driven disinformation running rampant on social media before US midterms, experts warn DevFeed: [AI-driven disinformation running rampant on social media before US midterms, experts warn]() Original publisher: [Read original article]() Author: George Chidi Published: 2026-10-27T10:00:10Z Content type: news Language: en Sources: [The Guardian | Technology]() Topics: [X (Twitter)](), [Instagram](), [Image]() Tags: [2016](), [2020](), [2023](), [accountability](), [acquisition](), [ads](), [agencies](), [ai](), [ai-artificial-intelligence](), [ai-generated-content](), [digital-media](), [disinformation](), [election](), [election-interference](), [elections](), [elon-musk](), [experts](), [facebook](), [internet](), [lies](), [media](), [meta](), [midterm](), [pennsylvania](), [social-media](), [technology](), [texas](), [us](), [us-midterm-elections-2026](), [us-midterm-elections-2026-us-politics-digital-media-us-news-social-media-media-world-news-election-i](), [us-news](), [us-politics](), [vote](), [world-news](), [x](), [youtube]() ### AI overview Experts say AI has made election disinformation cheaper to produce and easier to adapt around moderation, while social media companies have reduced trust and safety staffing and rely more on automated or community-based review. The excerpt traces false claims about Texas candidate James Talarico mutating across languages and a claim about a Pennsylvania campaign being amplified by an official DHS statement. It describes uneven platform rules: Meta requires disclosure for election ads using synthetic content and bans specific voting misinformation, but does little active moderation of political commentary; YouTube bars narrowly defined claims that could mislead or discourage voters, offers likeness-based removal requests, and promotes authoritative sources; X relies on Community Notes and has rules against voting misinformation and misleading synthetic media, whose enforcement experts question. Researchers warn that corrections may not cover evolving versions of a lie, and Frances Haugen argues platforms should examine distribution systems and feedback loops as well as individual posts. ### Source excerpt Social media companies including Meta have pulled back on efforts to protect voters from election-meddling lies Social media feeds are awash with an unprecedented flood of AI-driven disinformation ahead of November's US midterm elections, according to experts, who warn many posts are probably "slipping through the cracks". As tens of millions of Americans prepare to cast their vote, social media companies have scaled back their efforts to protect voters from lies meant to manipulate elections. Continue reading... ## CVE-2026-92885 - Qode Membership \<= 2.0.6.2 - Authentication Bypass via Social Login DevFeed: [CVE-2026-92885 - Qode Membership \<= 2.0.6.2 - Authentication Bypass via Social Login]() Original publisher: [Read original article]() Author: CVEFEED.io; wordfence.com Published: 2026-10-27T09:16:34Z Content type: news Language: en Sources: [Latest High/Critical Vulnerabilitiy Feed]() Topics: [Authentication](), [account takeover](), [Secure token management](), [conditionals]() Tags: [2026](), [account](), [account-takeover](), [administrator](), [ajax](), [authentication](), [authentication-bypass](), [cve-2026-92885](), [exploitation](), [facebook](), [graph](), [login](), [network](), [oauth](), [plugin](), [public](), [token](), [variable](), [verification](), [wordpress]() ### AI overview Qode Membership for WordPress versions up to and including 2.0.6.2 has a critical authentication bypass in the Facebook social-login AJAX action. A variable-name typo makes the OAuth token verification guard ineffective, leaving the authentication-cookie function reachable without credential or token validation. An unauthenticated attacker could log in as an existing account, including an administrator, by providing its login name in the response[email] parameter. Exploitation requires a valid qode_validate_facebook_login nonce, which the excerpt says is available on public pages using the [qode_user_login] or [qode_user_register] shortcode. The excerpt rates the issue CVSS 9.8. ### Source excerpt CVE ID :CVE-2026-92885 Published : Oct. 27, 2026, 9:16 a.m. | 1 hour, 28 minutes ago Description :The Qode Membership plugin for WordPress is vulnerable to Authentication Bypass leading to Account Takeover in all versions up to, and including, 2.0.6.2 via the qode_membership_check_facebook_user AJAX action. This is due to a typographic error in the Facebook OAuth token verification guard within the qode_membership_login_user_from_social_network function -- the guard condition tests an undefined variable ($check_social_useqoder instead of $check_social_user), causing empty() to always return true and rendering the Graph API token verification branch dead code, leaving wp_set_auth_cookie() reachable with no credential or token validation whatsoever. This makes it possible for unauthenticated attackers to log in as any existing WordPress account -- including Administrator-level accounts -- by supplying a known or enumerated administrator login name (rather than an email address) in the response[email] parameter, which the fallback path in get_user_by('login', $email) resolves to the matching account. Exploitation does require harvesting a valid qode_validate_facebook_login nonce, but this nonce is rendered on any public front-end page containing the [qode_user_login] or [qode_user_register] shortcode and is therefore obtainable without any prior authentication. Severity: 9.8 | CRITICAL Visit the link for more details, such as CVSS details, affected products, timeline, and more... ## Meta to sell up to $95B in stock to finance AI infrastructure buildout DevFeed: [Meta to sell up to $95B in stock to finance AI infrastructure buildout]() Original publisher: [Read original article]() Author: Maria Deutscher Published: 2026-10-26T21:07:07Z Content type: news Language: en Sources: [SiliconANGLE]() Topics: [Finance](), [AI Infrastructure](), [data centers](), [Nvidia]() Tags: [ai](), [ai-infrastructure](), [ai-infrastructure-buildout](), [data-center](), [facebook](), [finance](), [fundraising](), [hyperion](), [infra](), [infrastructure](), [llm](), [meta](), [muse-wonder-full](), [nasdaq-meta](), [news](), [public-offering](), [stock](), [the-latest](), [top-story-3](), [top-story-6]() ### AI overview Meta plans up to $95 billion in stock offerings to help fund its data center buildout. The initial transactions total $35 billion, including common stock and $12.5 billion in depositary shares that convert to common stock in 2029; Nvidia, AMD and Broadcom have expressed interest in buying up to $5 billion of shares. Underwriters may purchase additional shares, and a later at-the-market offering is expected to raise as much as $60 billion, starting early next year. The article describes four planned AI superclusters, including the 5-gigawatt Hyperion campus, which will use 10 new power plants. Meta is also using investment partnerships to finance projects and says it aims to reduce AI model hardware needs, improve inference infrastructure, and use off-peak computing capacity more effectively. ### Source excerpt Meta Platforms Inc. will sell up to $95 billion worth of stock to finance its data center projects. The fundraising initiative, which the company announced today, will kick off with a series of transactions cumulatively worth $35 billion. About two-thirds of that sum will come from the sale of common stock. Meta said Nvidia Corp., [...] The post Meta to sell up to $95B in stock to finance AI infrastructure buildout appeared first on SiliconANGLE. ## Meta plans $95bn stock offering to help cover AI expansion DevFeed: [Meta plans $95bn stock offering to help cover AI expansion]() Original publisher: [Read original article]() Author: Dan Milmo and Graeme Wearden Published: 2026-10-26T13:37:26Z Content type: news Language: en Sources: [The Guardian | Technology]() Topics: [Artificial Intelligence](), [Finance]() Tags: [ai](), [ai-artificial-intelligence](), [ai-companies](), [ai-infrastructure](), [ai-investment](), [ai-market](), [ai-startups](), [amd](), [announcement](), [business](), [cloud-computing](), [computing](), [datacentres](), [elon-musk](), [facebook](), [funding](), [fundraising](), [instagram](), [mark-zuckerberg](), [meta](), [meta-ai-artificial-intelligence-technology-sector-elon-musk-technology-startups-technology-busines](), [startups](), [stock](), [stock-markets](), [technology](), [technology-sector](), [technology-startups](), [whatsapp]() ### AI overview Meta plans to raise $95bn for continued AI infrastructure investment through two stock offerings: a $35bn underwritten public offering, with Nvidia, AMD and Broadcom expressing interest in buying up to $5bn, and a $60bn at-the-market sale of non-voting Class C shares. The structure lets Meta raise equity while preserving Mark Zuckerberg's voting control. The announcement came amid concern that highly leveraged neocloud data-center providers may struggle to service debt if AI demand does not grow quickly enough. Investors reacted differently to Meta and several such providers; an analyst cited Meta's ability to service debt and use equity raises. Musk warned that more than 200 neoclouds faced consolidation, while CoreWeave's CEO said some investors misunderstood the company. ### Source excerpt Tech firm announces two-part fundraising after Elon Musk warns hundreds of AI startups face collapse Mark Zuckerberg's Meta has announced plans for a $95bn (GBP 71bn) fundraising to cover its AI expenditure, even as Elon Musk warned that funding for smaller players in the sector was drying up. The owner of Facebook, Instagram and WhatsApp is a leading investor in AI and is ploughing multibillion-dollar sums into datacentres, the central nervous systems of the technology. Continue reading... ## Cambridge Analytica's fallout is still costing Meta, this time $25 billion in New Mexico DevFeed: [Cambridge Analytica's fallout is still costing Meta, this time $25 billion in New Mexico]() Original publisher: [Read original article]() Author: Alfonso Maruccia Published: 2026-10-23T16:01:00Z Content type: news Language: en Sources: [TechSpot]() Topics: [Finance]() Tags: [big-tech](), [business](), [cambridge-analytica](), [cambridge-analytica-scandal](), [consumer-protection](), [data](), [digital-advertising](), [facebook](), [fallout](), [fine](), [industry](), [internet](), [law](), [lawsuit](), [legal](), [meta](), [new-mexico](), [privacy](), [raul-torrez](), [social](), [tech-company](), [the-web](), [third-party](), [third-party-apps]() ### AI overview A New Mexico judge imposed a $25 billion civil penalty on Meta after a jury found Facebook had violated state consumer protection laws more than 43 million times in a case tied to the Cambridge Analytica scandal. The article says the violations involved unlawful third-party app access to user data, mishandled hate speech, and profiting from misinformation. The penalty was below the $219 billion maximum the judge said state law allowed, but above Meta's requested $3.45 billion cap and an attorney's proposed $35-40 billion range. Meta says the verdict could significantly affect its ability to operate in New Mexico and plans to appeal. An analyst cited in the excerpt notes the penalty is substantial, while Meta remains valued at nearly $2 trillion. ### Source excerpt Santa Fe District Court Judge Francis J. Mathew recently imposed a $25 billion fine on Meta, Facebook's parent company. The ruling follows a jury verdict in September that found the social network had deprived New Mexico residents of their privacy and their right to know how major technology companies use... Read Entire Article ## Facebook's misleading privacy claims land Meta a $25 billion penalty DevFeed: [Facebook's misleading privacy claims land Meta a $25 billion penalty]() Original publisher: [Read original article]() Author: Pieter Arntz Published: 2026-10-22T16:59:22Z Content type: news Language: en Sources: [Malwarebytes]() Topics: [online privacy](), [Security & Privacy](), [Access Control](), [least privilege]() Tags: [advertising](), [app-permissions](), [cambridege-analytica](), [cambridge-analytica](), [facebook](), [federal-trade-commission](), [meta](), [new-mexico](), [news](), [personal-information](), [privacy]() ### AI overview A New Mexico judge ordered Meta to pay $25 billion after a jury found Facebook made 26 misleading statements about privacy, hate speech and other matters. One key gap: post audience controls did not necessarily limit what outside app developers could access, which was governed by separate settings. The jury's 43,899,725 violation count estimated the audience for misleading statements, not the number of people whose data was taken. The judge reduced a potential penalty exceeding $219 billion; Meta disputes the case and expects to win on appeal. The penalty is directed to New Mexico's school system, not individual users. The article's precautions include limiting shared details, reviewing app permissions and tracking settings, scrutinizing connected apps, and remembering that restricting a post's audience does not prevent the platform from collecting or using activity data. It notes that settings cannot guarantee privacy. ### Source excerpt Users were misled about who could access their personal information on Facebook. Now Meta has been ordered to pay $25 billion. ## Meta ordered to pay $25 billion over Cambridge Analytica privacy case DevFeed: [Meta ordered to pay $25 billion over Cambridge Analytica privacy case]() Original publisher: [Read original article]() Author: Edward Komenda Published: 2026-10-22T09:20:08.594854Z Content type: article Language: en Sources: [The Proton Blog]() Topics: [online privacy](), [Security & Privacy]() Tags: [ads](), [ai](), [cambridge-analytica](), [data](), [facebook](), [meta](), [new-mexico](), [news](), [privacy]() ### AI overview A New Mexico judge ordered Meta to pay $25 billion plus 15% annual interest after a jury found more than 40 million violations of the state's Unfair Practices Act. The case centered on Facebook telling users they controlled their data while not explaining third-party access, including the Cambridge Analytica episode. Meta says the state did not show consumer harm and plans to appeal, which could take years. The penalty does not change Facebook's current data practices, and New Mexico law directs the funds to public education rather than affected users. The excerpt also describes other reported or alleged Meta privacy and platform controversies, while qualifying that Meta denied one allegation. ### Source excerpt A New Mexico judge ordered Meta to pay $25 billion for misleading Facebook users about their data. What the Cambridge Analytica ruling means. ## Meta Ordered to Pay New Mexico $25 Billion for Deceiving Users About Privacy on Facebook DevFeed: [Meta Ordered to Pay New Mexico $25 Billion for Deceiving Users About Privacy on Facebook]() Original publisher: [Read original article]() Author: Associated Press Published: 2026-10-22T07:26:54Z Content type: news Language: en Sources: [SecurityWeek]() Topics: [online privacy](), [Instagram](), [sensitive data]() Tags: [2](), [2016](), [2019](), [2021](), [about](), [accountability](), [added](), [ads](), [agent](), [ai](), [ai-agent](), [ai-assistant](), [award](), [data-protection](), [facebook](), [federal-trade-commission](), [fine](), [meta](), [new-mexico](), [privacy](), [privacy-compliance](), [users]() ### AI overview A New Mexico judge ordered Meta to pay $25 billion after a jury found more than 43 million consumer-protection violations tied to misleading statements about Facebook's handling of personal data. The case followed the Cambridge Analytica harvesting of data from roughly 87 million profiles. Meta says the state failed to prove deception and harm, calls the enforcement approach untenable, and expects to prevail on appeal. Jurors rejected separate claims that Facebook falsely described its removal of harmful content. The award exceeds the FTC's 2019 $5 billion fine; interest may accrue during appeal, but Meta could reduce the number of violations on which liability rests. New Mexico pursued the case separately after nearly all states and territories reached an $18 billion settlement with Meta that included stronger child-safety measures. ### Source excerpt The $25 billion award far exceeds a $5 billion fine the Federal Trade Commission levied against Facebook in 2019. The post Meta Ordered to Pay New Mexico $25 Billion for Deceiving Users About Privacy on Facebook appeared first on SecurityWeek. ## Meta hit with $25 billion fine for deceiving users on privacy DevFeed: [Meta hit with $25 billion fine for deceiving users on privacy]() Original publisher: [Read original article]() Author: James Farrell Published: 2026-10-22T01:10:20Z Content type: news Language: en Sources: [SiliconANGLE]() Topics: [Security & Privacy]() Tags: [c-j-mahoney](), [cambridge-analytica-fine](), [cambridge-analytica-lawsuit](), [cambridge-analytica-scandal](), [child-safety](), [facebook](), [facebook-fine](), [facebook-privacy](), [fine](), [judge-francis-mathew](), [lawsuit](), [meta](), [meta-fine](), [meta-lawsuit](), [meta-lawsuits](), [meta-settlement](), [new-mexico](), [news](), [policy](), [privacy](), [raul-torrez](), [settlement](), [social-media](), [social-media-addiction](), [the-latest](), [users]() ### AI overview A New Mexico jury found Meta liable for 43 million violations of the state's unfair practices act, concluding it misled users about data protection and use in a case tied to the Cambridge Analytica scandal. The violations carried penalties of up to $5,000 each, but Judge Francis Mathew reduced the state's $219.5 billion award to $25 billion, citing the scale and repeated, intentional nature of the conduct. The penalty grows 15% for each year it remains unpaid. Meta's chief legal officer said the company would appeal, arguing the ruling would make it nearly impossible to do business in New Mexico and improperly restrict online speech. ### Source excerpt Meta Platforms Inc. was ordered to pay $25 billion in penalties today after a jury found the company had misled its users on how it protects and uses their data. In what is one of the largest fines ever leveled against a tech giant, a jury in a Santa Fe, New Mexico, found the company [...] The post Meta hit with $25 billion fine for deceiving users on privacy appeared first on SiliconANGLE. ## Flaw in PHP computer language leads to shutdown of websites around world DevFeed: [Flaw in PHP computer language leads to shutdown of websites around world]() Original publisher: [Read original article]() Author: Robert Booth and Aisha Down Published: 2026-10-22T00:39:15Z Content type: news Language: en Sources: [The Guardian | Technology]() Topics: [PHP](), [Web](), [Artificial Intelligence]() Tags: [ai](), [ai-models](), [ai-tools](), [amazon](), [attacks](), [australian-signals-directorate](), [computing](), [cryptocurrency](), [cyber-attacks](), [data-and-computer-security](), [database](), [ddos](), [distributed-denial-of-service](), [exploit](), [facebook](), [internet](), [internet-data-and-computer-security-computing-programming-technology-world-news](), [patch](), [php](), [programming](), [shutdown](), [technology](), [vulnerability](), [websites](), [world](), [world-news]() ### AI overview A newly disclosed PHP vulnerability prompted temporary takedowns of government and other websites while maintainers worked to patch it. The US National Security Agency advised organizations unable to patch immediately to take systems offline; the UK's National Cyber Security Centre also urged checks for attacker-installed backdoors. Experts warned that public fixes and AI-assisted analysis could accelerate exploitation, leaving poorly maintained sites exposed for weeks or months. PHP issued a patch, but the excerpt does not specify the vulnerability's technical details or patching requirements. Researchers described different AI tools involved in finding the flaws, while the original researcher said they used publicly available models. ### Source excerpt Precautionary takedowns come as experts say bad actors could exploit vulnerability in language used by 70% of sites Websites around the globe have gone dark after a serious vulnerability was found in a computer language used to build much of the world wide web. Belgian, Dutch, Finnish, French and German sites were among those that governments took down temporarily as they scrambled to patch the critical weakness in PHP, a language used by almost 70% of websites. Continue reading... ## Meta ordered to pay $25bn to New Mexico in Cambridge Analytica data-privacy case DevFeed: [Meta ordered to pay $25bn to New Mexico in Cambridge Analytica data-privacy case]() Original publisher: [Read original article]() Author: Guardian staff and agency Published: 2026-10-21T23:33:34Z Content type: news Language: en Sources: [The Guardian | Technology]() Topics: [Security, Privacy and Abuse Prevention](), [Data Management]() Tags: [2016](), [cambridge-analytica](), [children](), [consent](), [consulting](), [consumer-protection](), [court](), [data](), [data-privacy](), [donald-trump](), [facebook](), [instagram](), [judge-francis-mathew](), [law](), [law-us](), [lawsuit](), [mental-health](), [meta](), [meta-us-elections-2016-donald-trump-facebook-cambridge-analytica-privacy-social-networking-technolog](), [new-mexico](), [payment](), [personal-data](), [privacy](), [social-networking](), [technology](), [us-elections-2016](), [us-news](), [us-politics](), [world-news]() ### AI overview A New Mexico judge ordered Meta to pay $25bn after a jury found 26 of 29 company and leadership statements misleading and counted more than 43 million violations of state consumer-protection laws. The case followed allegations that Cambridge Analytica obtained Facebook users' data without consent; New Mexico also accused Meta of misleading users about data access and harmful content. The judge said state law supported a penalty above $219bn but reduced it to $25bn. Meta had urged a $3.45bn cap, arguing the evidence showed it did not sell user data and no consumer was proven misled. The article also reports separate safety-related settlements and prior New Mexico penalties, including $375m in civil penalties and a further $567m for teen mental-health measures. ### Source excerpt Jury had found Meta misled consumers after Trump's 2016 presidential campaign obtained user data without consent A judge in New Mexico ordered Meta Platforms to pay $25bn on ⁠Wednesday after a jury found the company had misled consumers in a case arising from allegations that a political consulting firm involved in Donald Trump's 2016 presidential campaign obtained data from millions of Facebook users ⁠without their consent. "Facebook's wrongdoing - ⁠over the course of years and systematically broadcast to the public - goes to the heart of Facebook's core business practices," Judge Francis Mathew wrote in his order following the verdict. Continue reading... ## Meta Ordered to Pay New Mexico $25 Billion for Deceiving Users About Privacy On Facebook DevFeed: [Meta Ordered to Pay New Mexico $25 Billion for Deceiving Users About Privacy On Facebook]() Original publisher: [Read original article]() Author: BeauHD Published: 2026-10-21T19:00:00Z Content type: news Language: en Sources: [Slashdot]() Topics: [sensitive data]() Tags: [2016](), [ads](), [cambridge-analytica](), [ceo-mark-zuckerberg](), [consumer-protection](), [content-moderation](), [court](), [covid-19](), [education](), [facebook](), [meta](), [new-mexico](), [privacy](), [social-media](), [speech](), [users]() ### AI overview A New Mexico judge ordered Meta to pay $25 billion after a jury found Facebook liable for more than 43 million violations of state consumer protection laws tied to privacy claims and Cambridge Analytica's use of Facebook data for targeted political ads. Jurors assessed alleged deception 34 times and accepted the state's argument that some statements affected New Mexicans beyond platform users. They did not find the state had proved false claims about removing harmful content, including COVID-19 misinformation. Meta plans to appeal, arguing the law does not cover residents who were not Facebook users at the time; the state says collected funds will support education spending. The excerpt also reports Meta's attorneys said its policies have since changed and it removes 99% of content that violates its standards. ### Source excerpt A New Mexico judge has ordered Meta to pay $25 billion after a jury found Facebook liable for deceiving users about privacy during the Cambridge Analytica scandal. State Attorney General Raul Torrez called it "one of the largest, if not the single largest state verdict against a company in American history." While Meta plans to appeal the ruling, the funds that are to be collected will go toward state education spending. The Associated Press reports: Jurors in September found Meta liable for over 43 million violations of New Mexico consumer protection laws when a political consulting firm, Cambridge Analytica, harvested Facebook user data to generate targeted political ads for the 2016 election. Forty-eight other states settled with Facebook in a nationwide case, but New Mexico chose to take the social media company to court on its own. [...] During the trial, jurors assessed whether Facebook misled users about privacy and content moderation 34 separate times. State prosecutors, though, argued Facebook's statements deceived the entire population of New Mexico and extended beyond social media users. In the end, jurors agreed with this premise and found Facebook liable for 2.1 million violations on certain counts, bringing the total number of violations to over 43 million. Facebook's attorneys have argued in subsequent court filings that the state's consumer protection laws don't apply to people in New Mexico who weren't using the platform at the time. [...] In one of the few wins for the defense, jurors found the state didn't prove that Facebook made false claims about removing harmful content, including misinformation about the COVID-19 pandemic. The state said the company favored certain accounts and allowed for violent or inaccurate content to proliferate. Attorney Randi McGinn said in her closing argument for the state that Facebook profited off harmful content, which CEO Mark Zuckerberg denied during his video deposition. Zuckerberg said the company has to caref ## Google SynthID Detector Simplifies Watermark Detection and Expands Support to Other AI Providers DevFeed: [Google SynthID Detector Simplifies Watermark Detection and Expands Support to Other AI Providers]() Original publisher: [Read original article]() Author: Sergio De Simone Published: 2026-10-21T10:00:00Z Content type: news Language: en Sources: [InfoQ]() Topics: [synthid](), [generative-ai-tools](), [Image]() Tags: [ai](), [ai-generated-content](), [ai-ml-data-engineering](), [apple](), [audio](), [c2pa](), [cryptographic](), [detection](), [development](), [facebook](), [false-positives](), [gemini](), [global](), [google](), [google-ai](), [google-deepmind](), [media](), [metadata](), [microsoft](), [ml-data-engineering](), [news](), [nvidia](), [openai](), [partners](), [synthid](), [synthid-detector](), [synthid-watermarking-detector](), [watermark]() ### AI overview SynthID Detector at SynthID.com lets users upload images, audio, or video for checks across participating SynthID partners, including OpenAI, NVIDIA, and Kakao, with Apple support expected. This consolidates detection that previously required Gemini for Google-generated content or separate provider tools. It is not a general AI detector: it covers only participating providers, not Meta or Microsoft, whose watermarking differs. A negative result means no confident SynthID signal was found, not necessarily that content is human-made; extensive edits can weaken the watermark and cause false negatives. The text also describes C2PA metadata as a complementary way to record origin and editing history, while noting that uploads or screenshots can strip it. ### Source excerpt Google DeepMind announced the global availability of SynthID Detector, a tool designed to help identify AI-generated content, including media created using AI tools from SynthID partners such as OpenAI, NVIDIA, and Kakao, with Apple support expected soon. By Sergio De Simone ## Google has joined Amazon in attacking Nvidia's stronghold with AI app developers DevFeed: [Google has joined Amazon in attacking Nvidia's stronghold with AI app developers]() Original publisher: [Read original article]() Author: Julie Bort Published: 2026-10-20T18:38:24Z Content type: news Language: en Sources: [TechCrunch | AI News & Artificial Intelligence]() Topics: [Google](), [Nvidia](), [amazon](), [linux foundation]() Tags: [ai](), [ai-apps](), [ai-chip](), [ai-chips](), [ai-development](), [ai-development-tools](), [ai-industry](), [ai-models](), [amazon](), [apps](), [aws](), [cloud](), [developer-tool](), [developers](), [facebook](), [google](), [gpus](), [hardware](), [nvidia](), [pytorch](), [tc](), [tool]() ### AI overview Google's open-source TorchTPU lets developers run existing PyTorch apps on Google TPUs and Google Cloud through a configuration change, without rewriting them. Amazon's TorchNeuron similarly ports PyTorch apps to its Trainium chips and AWS. Amazon also plans to contribute code to the main PyTorch project to decouple features from Nvidia hardware. Google says it worked with Meta, Hugging Face, Inferact and RadixArk to support their development tools, and offers a TPU agent suite to help with migration and errors, plus updates to LiteRT for PyTorch apps. These tools lower migration effort, but the excerpt says they do not mean developers will automatically leave Nvidia. This overview is based on the supplied excerpt. ### Source excerpt Google has joined Amazon in releasing a developer tool that makes it easy for PyTorch users to move their apps off Nvidia GPUs. ## The Risk of AI-Powered Antidetect Browsers in Online Fraud and Influence Operations DevFeed: [The Risk of AI-Powered Antidetect Browsers in Online Fraud and Influence Operations]() Original publisher: [Read original article]() Author: matthewsu@checkpoint.com Published: 2026-10-20T14:17:23Z Content type: article Language: en Sources: [Check Point Research]() Topics: [online-fraud-prevention](), [Cybercrime](), [multi-accounting fraud](), [fake account fraud](), [account takeover](), [Payment Fraud](), [ai security](), [X (Twitter)]() Tags: [ai](), [ai-powered](), [ai-research](), [browsers](), [check-point](), [check-point-research-publications](), [detection](), [digital](), [discord](), [facebook](), [first](), [fraud](), [influence](), [instagram](), [introduction](), [key](), [linkedin](), [multiple](), [online](), [operations](), [operator](), [risk]() ### AI overview The report describes antidetect browsers as infrastructure for scaling account fraud and coordinated influence operations by hiding device identity and automating account activity. It examines Kameleo, AdsPower, and MultiLogin: Kameleo alters browser fingerprints and supports cookie injection; AdsPower offers large profile capacity, RPA templates, account warming, and an MCP server for LLM-generated content; MultiLogin combines isolated profiles with scripted automation and cloud phone profiles. The report links these capabilities to documented criminal use, including account takeover, phishing, fraudulent accounts, and influence campaigns, while noting the products also have legitimate uses. It says account and behavior checks may miss activity designed to appear benign, and recommends security teams examine network and infrastructure signals such as proxy use, WebRTC suppression, and anomalous cookie injection; platforms map these tools' bypass methods; and organizations review session monitoring and test antidetect scenarios. ### Source excerpt Key Findings Introduction Antidetect browsers conceal their users' digital identities, enabling a single operator or small team to simultaneously run thousands of isolated accounts across multiple platforms such as Facebook, Instagram, TikTok, YouTube, Twitter/X, Discord, LinkedIn, Reddit, and many more, without triggering the detection systems employed by those platforms. In this report, we highlight three [...] The post The Risk of AI-Powered Antidetect Browsers in Online Fraud and Influence Operations appeared first on Check Point Research. ## Instagram Modernized its UI with an AI-Native Architecture, Cutting Token Costs by 33% DevFeed: [Instagram Modernized its UI with an AI-Native Architecture, Cutting Token Costs by 33%]() Original publisher: [Read original article]() Author: Sergio De Simone Published: 2026-10-15T16:00:00Z Content type: news Language: en Sources: [InfoQ]() Topics: [Jetpack Compose](), [Vibe coding](), [React]() Tags: [a-b-testing](), [ai-assisted](), [ai-coding-agents](), [android](), [bugs](), [development](), [facebook](), [feature-flags](), [instagram-compose-migration](), [jetpack-compose](), [migration](), [mobile](), [news](), [performance]() ### AI overview Instagram Direct migrated its Android UI from Android Views to Jetpack Compose, shaping the codebase so AI coding agents can work with established patterns and architectural boundaries. The engineers report 50% less UI code, 33% lower token costs per agent session, 35% less agent execution time, and 32% fewer engineer-agent exchanges. They evaluated performance using production metrics and A/B tests as screens moved to Compose. ### Source excerpt Instagram Direct migrated its Android UI from the legacy Android View system to Jetpack Compose, creating an AI-native UI codebase that is easier for AI coding agents to understand, modify, and maintain. According to Instagram engineers, the migration reduced UI code volume by 50% and token costs per AI agent session by 33%. By Sergio De Simone ## Meta Extends Confidential Computing to AI Glasses with Private Processing DevFeed: [Meta Extends Confidential Computing to AI Glasses with Private Processing]() Original publisher: [Read original article]() Author: Leela Kumili Published: 2026-10-14T13:57:00Z Content type: news Language: en Sources: [InfoQ]() Topics: [Confidential Computing](), [Artificial Intelligence](), [Cryptography]() Tags: [agentic-ai-architecture](), [ai](), [ai-glasses](), [ai-ml-data-engineering](), [ai-security](), [architecture-design](), [attestation](), [cloud](), [confidential-computing](), [data-privacy](), [development](), [facebook](), [meta-private-process-ai-glasses](), [ml-data-engineering](), [news](), [privacy](), [storage](), [trust](), [vm](), [wearable-tech]() ### AI overview Meta is extending its Private Processing architecture to AI glasses, using confidential virtual machines and hardware trusted execution environments to handle sensitive AI workloads and persistent user context in the cloud. The design includes remote attestation, anonymous credentials, OHTTP relays, and encrypted storage. Meta says the system is not yet available on its glasses and has not announced an availability timeline or initial supported features. ### Source excerpt Meta has extended Private Processing to AI glasses, using confidential VMs, trusted execution environments, remote attestation, anonymous credentials, OHTTP routing, and TEE-based storage to process AI workloads and persistent user context in the cloud. By Leela Kumili [Next page]()