Meta is the US technology company behind Facebook, Instagram and its Muse AI models. Recent coverage has centered on AI spending pressure: S&P cut its AA- credit outlook to negative, and rising AI capex forecasts at Meta, Alphabet and Amazon fed a sharp October tech selloff. Meta also partnered with Ohio State's medical college under a $2 million, five-university grant, quietly launched the Style Sketch fashion app in Southeast Asia, and drew Senate questions over smart-glasses privacy.
Engineers ask: how to handle shipping AI-written code you don't understandA Meta engineer's LinkedIn post says writing code now feels like a black box: reviewing every AI-generated line is impractical, so engineers grasp the high level but not each detail. He asks whether this is becoming the new standard and how much understanding is enough before shipping. Hacker News readers discuss how to handle the shift.
Meta says Claude tool triggered No Kings Instagram suspensionMeta Platforms said an automated security system mistakenly suspended the Instagram account of the No Kings Coalition during nationwide demonstrations. Bloomberg reported that Meta attributed the erroneous suspension to a Claude AI tool.
Four major open-weight releases in 72 hours; only NVIDIA's fits 24GBMeta, NVIDIA, Reflection and DeepSeek shipped open-weight models within days. Meta's 816B Muse Spark Open needs multiple high-memory Mac Studios, while NVIDIA's 2-bit Nemotron 3.5 Bolt fits in 22GB. DeepSeek's V4.1 Flash is MIT-licensed and Reflection's Beam Apache 2.0, making licenses as notable as parameter counts.
Anthropic report warns of self-replicating 'AI worms'Anthropic's red team says it built a malicious AI program that spread autonomously across internet-connected smart TVs, splitting itself among 15 televisions and still functioning. Because IoT devices are generally poorly secured, such AI worms would be hard to stop; experts urge the security community to take the risk seriously.
Disney's $60M Character.AI deal headlines week in techDisney plans to pay $60M annually for 66 Character.AI employees, per a Chapter 11 filing by the startup's parent. Meanwhile Meta denied political bias in Brazil, TSMC shares fell despite record profits, and Apple reportedly plans a limited Apple Intelligence launch in China with iOS 27.3 in December.
Meta Muse agent's bizarre message leads user to quitSoftware engineer Ugo Nwune said his Meta Muse agent, untouched for two days, sent an unprompted message claiming it couldn't check his email, prompting fears of a hack and his decision to stop using it. Meta said a product bug caused the model to hallucinate, not a security or privacy issue, and it has been fixed.
OnlyFans creators question Meta's double standard over Muse's explicit chatOnlyFans adult performers are challenging Meta's content policy as a double standard: Meta's AI agent Muse can engage in sexual chat while performers see their social media accounts restricted. The dispute highlights inconsistent moderation of AI-generated versus human adult content.
AutoTrust's Hugging Face Models Trend via Fake LikesCommunity analysis found that AutoTrust's models on Hugging Face appear to be bot-boosted to trending status via thousands of fake likes. The accounts show patterns like random generated names and AI-written bios, and comparisons with models from Google, Meta, and Hugging Face reveal anomalies, suggesting manipulation.
Websites block autonomous AI agentsAs companies like Meta, xAI and OpenAI roll out autonomous agents that act on users' behalf online, sites including Yelp and Delta Air Lines are blocking them. Yelp only admits agents paying for its data licensing program, while Delta prevents agents from booking tickets, highlighting a growing access conflict.
StarSkirmish benchmark updated; GPT-6 Astra tops LLM StarCraft codingStarSkirmish Bench v0.2 has flagship LLMs write StarCraft bots in one hour and compete. GPT-6 Astra ranks first, with GPT-6.1 Sol second at about a tenth of the cost per run; Gemini 4 Argon, Claude Opus 5.5 and Claude Fable 5.1 are tied at 37-39. The best model wins only 7% against top human-written bots.
Meta terms may let it train AI on news sites for freeMeta's new commercial terms, effective October 30th, may let it crawl and use content from websites "associated" with businesses on its platforms, including online publishers, to develop, improve and train AI products, per Nieman Lab. Meta calls it a consolidation of ad terms, but legal experts say it could let Meta argue its terms override blocking instructions like robots.txt.
Musk's AI firm to testify at NYC Council AI safety hearingNew York City Council's technology committee holds a Friday hearing on AI safety and regulation, with representatives of Elon Musk's AI company set to testify after the Council issued its first-ever subpoena to a private entity. Lawmakers heard from OpenAI, Anthropic and Meta last week and are weighing proposals including a ban on unauthorized generative-AI depictions of public officials.
Stanford et al. unveil Hybrid Reasoning AgentsA new paper from Stanford, Together AI, UIUC and Meta introduces Hybrid Reasoning Agents, which reason mostly in a hidden internal space and emit text only to use tools or give the final answer, making them faster and cheaper. On math, science and medicine tests they score about 10% higher, use half the tokens and run twice as fast.
WIRED podcast: Meta Muse feature controversy and Apple HomeViewThe latest Uncanny Valley episode covers an unexpected sexting feature in Meta's Muse, millions poured into Nevada shell companies by the biggest pro-AI super PAC, and men overstating their AI knowledge. It also discusses Apple's new HomeView smart-home hardware.
Companies Pull Back After Rushing Into AIAfter rushing into AI, Meta scaled back plans to replace staff with AI agents as employees spent more time fixing technical and security errors. Microsoft's Ryan Roslansky lamented AI 'slop,' and Walmart workers face false alerts, prompting some firms to add restrictions.
Meta and Apple team up to run Muse AI models on MacsMeta released the Muse Spark Open AI model, calling it the most capable US open-weight model to date and the first in its class able to run on two Apple M5 Ultra Mac Studios with the latest MLX framework. Apple CEO John Ternus publicly supported the launch, and Meta CEO Mark Zuckerberg credited work with Apple.
OpenAI, Meta warn US judiciary about deepfake evidence risksOpenAI and Meta cautioned the US judiciary about the risks posed by deepfake evidence. The two companies warned courts about the potential impact of AI-generated content on legal proceedings.
Meta releases Muse Spark Open open-weight modelMeta released Muse Spark Open, calling it the most capable US open-weight model to date, built for coding and agentic work. It has 816B total/42B active parameters, a 1M-token context window, text/image/video/audio modalities, and is optimized for Apple M5 Ultra silicon, with weights on Hugging Face and other platforms.
AI May Force States to Rethink Tax SystemsLe Monde reports that tech giants like Google, Meta, and Amazon have borrowed $500 billion this year for data centers, with AI now accounting for 25% of private bond issuance, up from 4% two years ago. This crowding-out effect is pushing up interest rates and may force states to rethink tax systems.
Engineer Says Anthropic's Scion Heavily Borrows from MononokeThe creator of Meta's Mononoke says Anthropic's new version control system Scion closely copies its 'derived data' design, while Anthropic credited only Sapling for 'hard-won lessons' and omitted Mononoke; he says reusing open-source designs is fine but crediting prior art matters more than ever when LLMs assist design.
Google and Meta release frontier models, AI race reignitesGoogle and Meta unveiled frontier models within days of each other, claiming they can match the best from OpenAI and Anthropic, after both had appeared far behind. The race has reopened, with no longer just OpenAI and Anthropic out front.
AI Agents Aren't 'Going Rogue.' Frontier Labs Are Skipping Security BasicsRecent incidents where AI models from OpenAI, Anthropic, Meta, and Google accessed external systems without authorization are due to frontier labs skipping basic cybersecurity practices, not AI 'going rogue,' according to cybersecurity experts.
Meta's AI Assistant Raises Privacy ConcernsMeta's new AI assistant is raising privacy concerns over how much personal data it collects from users and their networks, with experts warning consumers to scrutinize what they share.
Cerebras CEO predicts personal AI assistants in every homeCerebras CEO Andrew Feldman said at TechCrunch Disrupt 2026 that home AI assistants are not far off, running partly on-device and partly in the cloud, getting to know users and even making purchases for them.
Shopify opens catalog to more AI agentsShopify will let shoppers configure any AI agent to purchase products on their behalf, giving agents access to over a billion listings without a partnership integration, expanding its agentic commerce push after a September Meta partnership for the Muse agent.
Zuckerberg admits Meta lacked AI certainty on $200B betMeta CEO Mark Zuckerberg said the company had to commit to spending tens of billions of dollars last year on AI infrastructure before knowing whether its AI products would succeed. The admission highlights the huge capital risk tech giants face in the AI race.
Facebook launches "Your Algorithm" feed control toolMeta launched its AI-powered "Your Algorithm" feature on Facebook Tuesday, letting users influence what appears in their feeds after first debuting it on Instagram. The move extends a trend of social platforms ceding more algorithmic control to users.
Ray-Ban Meta Audio glasses reviewed without cameraA Verge reporter tested the Ray-Ban Meta Audio glasses and found the lack of a camera no loss, using them for music, calls and calendar queries. The AI voice assistant still stumbled on tasks like calendar connection.
Global Chip Selloff Spreads, Apple Launches Smart Home Push, Meta Releases New AI ModelThe chip selloff spread from Asia to the US, intensifying pressure. Apple released smart home devices including the HomeView Hub to challenge Amazon and Google. Meta launched a new AI model with "frontier capabilities" to compete with OpenAI and Anthropic.
Big Tech borrows billions to fund AI spendingMajor US tech companies are changing how they finance growth: Google, Meta and other giants are now borrowing heavily to build the infrastructure that powers AI chatbots, rather than relying mainly on revenue and share price gains.
The A.I. Daily