Meta co-founder Mark Zuckerberg said its newest models are a step toward “personal superintelligence.” Meta has unveiled two new artificial intelligence models, Muse Wonder and Muse Wonder Full, which it called its most capable models to date and a step toward “personal superintelligence.” Meta on Monday said the models introduce two capabilities Meta calls Active Memory and Active Focus. The first carries relevant information across conversations, and the second helps the agent sustain work on difficult tasks. Independent evaluation platform Artificial Analysis scored Wonder Full at 56 on its Intelligence Index and the standard Wonder at 52, both at maximum reasoning effort. Read more
The incident reportedly stemmed from a misconfigured testing environment, adding Meta to a growing list of AI firms whose models have escaped evaluation sandboxes. Meta has become the latest major AI company to disclose that one of its models hacked another company’s systems during testing, following similar incidents involving Anthropic and OpenAI. The model involved Meta’s Muse Spark 1.1, which launched in July, according to The Information, citing sources. The issue reportedly stemmed from a misconfiguration by Irregular, an artificial intelligence security testing and red-teaming firm, which inadvertently gave the model internet access during an evaluation. The model “exploited a security vulnerability in a third-party service, in a manner similar to previously reported instances with other companies,” Meta told Reuters in a statement. Read more
Approximately 49,000 workers were laid off in 2026 as companies adopted a more AI-reliant business model. Meta has reportedly begun cutting staff in Singapore as the company executes a plan to reduce employee numbers by 8,000 to lean more heavily on AI. Emails were sent out at 4 a.m. Singapore time to affected employees, and staff in the US and Europe were also expected to be notified that morning, Bloomberg reported on Tuesday, citing people familiar with the matter. Meta’s engineering and product teams are expected to be hit hardest. Meta is one of several Big Tech firms cutting staff while investing heavily in AI infrastructure in an effort to streamline operations and reduce costs. An estimated 49,135 layoffs have occurred at US companies in 2026 as a result of AI integration. Read more
Elizabeth Warren asked the Meta CEO to provide details on a stablecoin integration to the platform, a week after a small rollout to creators in Colombia and the Philippines. Massachusetts Senator Elizabeth Warren called on Meta CEO Mark Zuckerberg to answer questions about the company’s stablecoin integration in 2026, signaling concerns about guardrails. In a Wednesday letter to Zuckerberg, Warren said Meta’s lack of transparency regarding its stablecoin was “deeply troubling,” given the company’s previous plans to roll out Libra, a global stablecoin proposed in 2019 that was later rebranded to Diem. The senator said Meta’s plans were necessary for Congress to understand, given the US government’s efforts to pass a digital asset market structure bill with implications for stablecoin issuers. Read more
The latest stablecoin move comes years after Meta first dipped its toes into stablecoins through Libra in 2019, before scrapping the plans in 2022 due to regulatory pushback. US tech giant Meta has launched USDC payouts for creators on its platforms in the Philippines and Colombia, with plans to expand to additional markets. Creators who opt into the service will receive payments directly into crypto wallets on the Solana and Polygon blockchains. However, Meta does not offer a built-in conversion option, so an external exchange is required to convert USDC into fiat currency, according to the announcement on Wednesday. The service is currently available only to select creators in Colombia and the Philippines, but Polygon said Wednesday that the stablecoin rollout is expected to expand to more jurisdictions soon. Read more
The Meta co-founder is reportedly working on a personal AI agent to bypass management layers as Meta pushes employees to adopt agentic tools. Meta CEO and co-founder Mark Zuckerberg is reportedly building an AI agent to help handle his work in managing the company amid a company-wide push for employees to adopt agentic tech. According to a report from The Wall Street Journal on Sunday, citing sources close to the matter, Zuckerberg’s AI agent is still in development but already being used to help the CEO speed up information retrieval. Instead of going through multiple layers of people or teams to get the required information, the agent has been retrieving the information directly. Read more