Muse Wonder is Meta Superintelligence Labs' frontier model family. Meta says the new specialized models carry two of its capabilities, Active Memory and Active Focus, into the Muse app and Meta AI. The same day Alexandr Wang announced Spaces in Muse, pages Muse builds and keeps up to date from a chat request.

AI at Meta said on Friday that Muse and Meta AI now run on specialized models trained using Muse Wonder, its frontier model family. The goal, according to Meta, is to bring some of Muse Wonder's capabilities for long, complex tasks to everyday users rather than keeping them in the flagship model.

Meta highlighted two capabilities. Active Memory is meant to recognize when earlier information in a conversation still matters and to tell relevant details apart from stale ones. Active Focus is meant to help the model recover from setbacks or find alternatives when constraints change during a task.

Meta gave two numbers for Muse: a 2.4x reduction in forgetting information critical to the user's goals, and a 67% improvement in getting unstuck as constraints evolve. In Meta AI, the company says the update helps the assistant break complex problems into steps, ask sharper follow up questions and adapt when the user changes direction. These figures come from Meta's own testing and have not been independently measured.

Alongside the model update, Alexandr Wang announced Spaces in Muse. A user describes what they want in chat, Muse builds a page on that topic and keeps it current, and the page can be pinned to the navigation bar.

Watch for independent tests of the memory and focus claims, and for whether Meta publishes benchmark results for the specialized models beyond its own blog figures.

Open questions include which Muse Wonder capabilities stay exclusive to the flagship model, how fast the update reaches every Meta AI surface, and how Spaces compares with similar live pages from OpenAI and Google. Usage numbers for Muse and Meta AI after the update would show whether the changes bring people back.

zerohedge posted the wire headline on Friday afternoon. Bloomberg reported, as relayed by Rohan Paul, that investors wired the money that day and that this is the first tranche of a larger round. The close came as SoftBank, one of OpenAI's largest backers, kept falling.

A headline carried by zerohedge on Friday said OpenAI closed $11 billion of a new funding round at a $1.25 trillion valuation. Rohan Paul, citing Bloomberg, described it as the first tranche of the round and said investors had wired the cash that day, with the rest expected later. OpenAI had not commented on X at the time of the scan.

The timing stands out. On Thursday and Friday AI stocks sold off after Democratic midterm wins raised fears of tighter rules on data centers and AI. SoftBank, which holds a large OpenAI stake, fell another 10.8% on Friday. Closing billions in fresh money at a higher mark during that selloff suggests private investors still see OpenAI's growth as worth paying for.

The same day OpenAI's CFO Sarah Friar told CNBC the midterm results did not dramatically change the company's plans, according to posts relaying the interview. The size of the full round and the names of the investors in this tranche were not part of the headline.

Watch for the full round size, the investor list and whether later tranches close at the same price.

Also watch how this mark compares with Anthropic's IPO target of $1.5 trillion, and whether public market weakness in AI names pushes later investors to ask for better terms. A confirmation from OpenAI itself, with the share of the round that is now closed, would settle how much of the headline is final.

unusual_whales relayed a Washington Post report that the House is laying the groundwork for an investigation. The report concerns the use of xAI's Grok in planning the military operation to capture Nicolás Maduro. Andrew Curran separately relayed WaPo reporting that Democrats plan investigations touching SpaceXAI, Meta and Apple.

According to the Washington Post, as relayed by unusual_whales, the House is laying the groundwork to investigate the Trump administration's use of Grok, xAI's chatbot, to plan the military operation that captured Nicolás Maduro. The post drew about 456K views within hours.

The report lands as Democrats prepare to take control of Congress after the midterms. Andrew Curran relayed further WaPo reporting that Democrats are preparing congressional investigations into SpaceXAI, Meta and Apple. Reuters separately reported that Democrats plan a flood of AI bills once they hold the majority.

No investigation has been formally opened, and the details of how Grok was used have not been confirmed by the White House, the Pentagon or xAI in the posts scanned. The story is a report about planned oversight, not a finding. It is also one of several signs this week that AI companies close to the administration will face more scrutiny from the next Congress, alongside planned bills on data centers, AI safety and liability.

Watch for formal letters or subpoenas once the new Congress is seated, and for any response from xAI, the White House or the Pentagon.

A key open question is what role a commercial chatbot actually played in military planning, and whether the probe leads to rules on government use of AI models. Also watch whether other AI companies with government contracts, such as OpenAI, Anthropic and Google, face similar questions about how their tools are used by the military and intelligence agencies.

Google DeepMind shared research on using AI agents, guided by humans, to stress test encryption. Its VP of Science Pushmeet Kohli explained the work in a thread. The weaknesses concern legacy 1024 bit RSA systems, a key size already considered outdated.

Google DeepMind said on Friday that it is sharing new research showing how AI agents can accelerate the science of stress testing encryption. Guided by human researchers, its models uncovered two major weaknesses in legacy 1024 bit RSA cryptosystems, according to the lab.

RSA at 1024 bits has been considered too weak for new systems for years, but it remains in older hardware and software. Research that finds concrete weaknesses in those deployments helps security teams decide what to replace first.

DeepMind framed the work as human guided, meaning researchers directed the agents rather than the models working alone. The announcement came a day after OpenAI introduced GPT 6 Cyber for security research, which shows labs racing to show their models doing real security work.

The posts did not claim a break of modern RSA key sizes. Details of the two weaknesses are in Pushmeet Kohli's thread and the linked research.

Watch for outside cryptographers to review the findings and for vendors still using 1024 bit RSA to respond.

The broader question is how quickly AI driven cryptanalysis moves from legacy systems to current standards, and how labs handle disclosure when their models find real flaws. Expect more labs to publish similar results as AI security work becomes a main way to show what new models can do.

Watcher.Guru relayed the Financial Times report that Anthropic plans to IPO before Thanksgiving at a $1.5 trillion valuation. The FT headline says Anthropic will press ahead with that target. On Nov 5 the WSJ reported that big investors were skeptical of the price.

The Financial Times reported on Friday that Anthropic will press ahead with its IPO at a $1.5 trillion valuation target, and Watcher.Guru relayed that the listing is planned before Thanksgiving. Anthropic has not confirmed the timing on X.

The report follows Thursday's WSJ story, covered in the Nov 5 deck, that large investors were pushing back on the $1.5 trillion price after the market drop. Pressing ahead suggests Anthropic believes demand is strong enough despite the selloff in AI stocks after the midterms.

If it goes ahead on that schedule and at that price, it would be the largest AI company to go public. The same afternoon OpenAI reportedly closed $11 billion of a private round at $1.25 trillion, so the two leading labs are now being priced in very different ways. One is raising privately in tranches, the other is preparing to sell shares to the public within weeks. Details on how many shares Anthropic would sell, and on which exchange, were not in the posts scanned.

Watch for a public filing, the final price range and how much stock Anthropic sells.

The open question is whether public investors accept $1.5 trillion while AI and data center names are under pressure, or whether the price moves before the listing. Also watch how OpenAI's private round at $1.25 trillion shapes the debate over which lab deserves the higher price, and whether other AI companies line up listings behind Anthropic if it goes well.

OpenAI built the benchmark with Ligo, a biotech company focused on enzyme design. Its 100 tasks come from real experiments and test whether agents can read raw lab data, spot errors and recognize unsupported conclusions. OpenAI says it will help track progress for science and for biosafety.

OpenAI introduced BioBasicsBench on Friday, a benchmark for how AI agents handle the routine analyses of wet lab science. It was developed with Ligo, a biotech company focused on enzyme design, and draws on 100 tasks from real experiments.

The tasks test whether an agent can interpret raw lab data, identify errors and recognize when the evidence does not support a conclusion. Posts relaying the release describe inputs such as gel images and instrument readouts rather than clean text questions.

OpenAI framed the benchmark as a way to see where models still need to improve and to track what gains mean for scientific progress, biosafety and biosecurity. That dual framing matters because better lab skills can help both research and misuse.

The launch fits a week of science and security releases from OpenAI, including GPT 6 Cyber on Nov 5. Scores for specific models were not in the post, and OpenAI did not say how its own models performed.

Watch for published scores across frontier models and for outside labs to adopt the benchmark.

Also watch how OpenAI uses the results in its biosafety evaluations, and whether Ligo or others release more tasks. A key question is whether models that score well here also cut real lab mistakes, and how OpenAI balances wider access to strong lab skills against biosecurity risk. Results from other labs on the same tasks would show whether OpenAI's models lead.

Reuters reported the plans after Democrats won control of Congress in the midterms. A Reuters reporter said on air the focus will include data center regulation and AI safety. Markets sold off sharply on fears of tighter AI rules.

Bloomberg reported that Microsoft is winding down an arrangement to rent AI computing from Oracle. Posts relaying the report say the 2023 contract was built around Bing and older Nvidia chips. Bloomberg ties it to a wider review of billions in cloud partnerships.

zerohedge posted the headline that Google bought Crusoe's Wyoming data center site. Bloomberg reported Google opted to take on the southeast Wyoming project itself. On Nov 5 Google said it is building a $17B data center in Wyoming.

OpenAI partnered with the Frontier Model Forum and its member firms on the framework. It aims to identify when bypassing an AI system's safeguards could enable serious cyber harm. OpenAI's Sasha Baker called it a step toward a common method for comparing jailbreak severity.