Replit announced on X that TikTok Ads MCP is now available in Replit, letting users connect a TikTok Ads account and bring advertising workflows into the workspace where they build. The post only links to the integrations page and gives no detail on supported tools, regional availability or pricing.
California's combat-sports regulator ordered San Francisco startup Robot Entertainment Kombat (REK) to stop staging or promoting boxing or MMA events involving a human without approval, after REK put content creator Frankie LaPenna in a cage against a reported 6-ft. EngineAI T800 rebranded as "Rekbot" and operated by a human in a VR headset; the founder posted the cease-and-desist as publicity, noting robot-vs-robot events are not covered. The same roundup reports Boston Dynamics gave Atlas new four-finger hands with 13 degrees of freedom per hand (up from seven) that can grip and trigger power tools, and that FieldAI signed a term sheet for a $700M round at a $10B valuation.
U.S. President Donald Trump announced the creation of the Super Intelligence Force, led by Director of National Intelligence Jay Clayton and other administration officials. The task force is expected to submit a report within 120 days on AI-related risks and opportunities while developing responses to social threats and avoiding excessive regulation, according to its charter.
The VA-Bench paper evaluates 12 multimodal models in closed-loop robot tasks spanning observation, reasoning, action, and revision. The source reports that leading models achieved near-perfect target localization and action-semantic scores, but complete-task success remained around half, with online correction, fine-grained control, active perception, and bimanual coordination as major weaknesses.
Meta is reportedly cutting a group of employees recruited from AI safety startup Virtue AI in June, with spokesperson Andy Stone telling Semafor that the partnership did not develop as expected because of “clashing work styles.” The report does not confirm whether Dawn Song and all other Virtue AI members have left, and Virtue AI had not responded at publication time.
Enterprise predictive analytics is shifting toward systems that can act on their own forecasts, according to an MIT Technology Review Insights report. The report highlights real-time training, generative AI, deep learning, and unstructured data as drivers of more autonomous decision-making, while stressing the need to keep actions aligned with business intent.
ZDNET editor David Gewirtz handed a 30-year-old Macromedia Director icon file to Claude, which extracted 2,000-plus PNGs in minutes, then used Claude Code and Claude Desktop to build the Mac app Icon Gallery '97: 30th Anniversary Edition. He submitted it to Apple on Sept. 27 and it was accepted and live on the App Store four days later. He stresses this was not one magic prompt: he issued 502 directives to Claude Code across four days of active interaction.
Elon Musk said SpaceXAI may be renamed SpaceXSI after a user suggested the new name, saying the change would happen. The company, formerly xAI, was renamed SpaceXAI after SpaceX acquired it earlier this year, but no effective date or new branding has been announced.
An analysis uses OpenAI’s reported safety and culture disputes to argue that increasingly autonomous AI agents expose weaknesses in corporate governance. It recommends defining permissions before deployment, monitoring actions in real time, and assigning clear human responsibility for high-risk tasks.
An industry review says AI has rapidly expanded short-drama production, with translation, marketing assets and distribution currently seeing more adoption than fully AI-generated series. DataEye figures cited in the review show 221,900 AI dramas and animated dramas launched on Douyin in the first half of 2026, but only 1,055 exceeded 100 million views, while high user-acquisition costs and platform concentration continue to limit profitability.
Simon Willison re-ran an experiment Colin Frasier first tried two years ago with GPT-4o: have a model compute large sums but return the answer in words. He ran it on local hardware (a DGX Spark), using a Codex Remote session (GPT-6 Astra) to write and execute the test with Qwen3.8-27B-Q4_K_M.gguf. With reasoning disabled he sampled 30 attempts per cell; with reasoning enabled each run took so long that he used one sample per pair, and the model got 167 of 169 attempts right.
OpenAI Foundation has committed $25 billion to philanthropy, with initial priorities including health and disease research and AI resilience. It has announced more than $125 million for public health data and more than $100 million for an AI-for-Alzheimer's program, while requiring funded datasets to be broadly available to researchers. The article notes that the foundation's close financial and governance ties to OpenAI Group PBC could create perceived conflicts of interest, though it cites no public evidence of improper influence.
Manus Community shared a short clip featuring @hidecloud discussing the thinking behind Manus 2.0. A longer conversation is expected later, alongside upcoming Manus events in several cities and online.
A product listing for Vresk has appeared on Product Hunt, describing itself as "one AI workspace that picks the best model for each task." The listing offers no further detail on which models are supported, pricing, or general availability.
Former Anthropic researcher Jacob Coxon is expected to testify at a New York City hearing on artificial intelligence, according to people familiar with the matter cited by Bloomberg. Coxon previously warned that AI could potentially cause human extinction by the end of this decade and accused Anthropic and OpenAI of risking lives.
SKF has released an advertisement featuring an AI-generated likeness of late Hollywood actor Greta Garbo. The company says the visual character was created from text prompts without archival photos or footage, while her voice was trained from one authorized archival film recording and refined with help from a human actor.
MIT Technology Review has made the full EmTech Future 2026 program available on demand. The event examined how AI intersects with biology, infrastructure, manufacturing, science, energy, quantum technology, and robotics, with sessions including remarks from Google Research’s Yossi Matias and Google Quantum AI’s Hartmut Neven.
This Chinese-language essay describes the excitement and anxiety around AI: it cites a Pew Research Center survey said to cover 37 countries in September 2026, in 34 of which more people feared AI would cut jobs than expected it to create them, while a 2025 International Labour Organization and NASK study is said to conclude AI is more likely to change job content than eliminate roles. The author argues that polishing old skills, being first with new tools, or simply producing more are not durable answers, and proposes turning AI collaboration into personal accumulation: state your expectations first, explain reasons when comparing outputs, and expose your standards to outside feedback. The piece adds that better judgment and taste still do not guarantee security, since efficiency gains can be absorbed by new organizational requirements, and that the aim should be to keep some of that efficiency for yourself.
Replit's account promoted a new Times Tech podcast episode featuring Replit CEO Amjad Masad (@amasad), who discusses vibe coding, the lost joy of building with computers, and why AI doomerism misses the accountability question. His central question: if AI agents can act on our behalf, who is responsible when they go too far.
QbitAI reports that OpenAI has committed to a 28-day window: users can try an update if there is an improvement, or reset if there is not. The available excerpt provides no further details about the commitment or the product involved.