Meta / Llama

Order 837 took effect July 1, extending jurisdiction to individual residents, treating cross-border personnel services as technology exports, and creating an outbound investment security review. The trigger: China's NDRC blocked Meta's ~$2B acquisition of Manus by piercing its Singapore domicile and tracing tech, talent, and IP back to China. Domestic firms now face five stacked reviews; VIE-structured overseas IPOs still work—MiniMax and Zhipu listed in Hong Kong in 2026. Offshore-incorporated startups lose jurisdictional immunity, and Article 22 erects a data wall that makes cross-border litigation a no-win bind. Chinese citizens employed abroad aren't covered by the new rules, but founders holding equity fall under Article 33—yet no filing channel exists as of July 12. The analysis is based on statutory text and law firm interpretations; zero enforcement cases so far.

Meta launched Muse Image on Instagram Tuesday, letting anyone use public account content to generate AI images with users opted in by default. After swift privacy backlash, Meta admitted it “missed the mark” and pulled the feature. Sag-Aftra and Privacy International both criticized it. Meta says the intent was a creative tool; the post doesn’t say if it will return as opt-in.

Meta removed the Muse Image feature on Instagram less than a week after launch. It let users @-mention any public account to use their photos as AI image-generation references without notifying them. Meta said in a blog post the feature “missed the mark” and is no longer available. TechCrunch had published a guide on how to opt out before the reversal.

Meta launched an Instagram feature on July 8 that let users create AI deepfakes of public accounts via DM, then shut it down two days later after backlash. The tool, called Muse, worked by messaging @MetaAI with 'imagine me as [public account]' to generate a styled fake image. A Meta spokesperson confirmed the feature is off but didn't explain why. The post doesn't disclose usage numbers or any actual harm cases. This reads more like a quick trial that got pulled after pushback, not a formal product rollout.

Zuckerberg denied Meta has excess compute, saying all resources run at full capacity. But he admitted market bids are so high that renting out some AI infrastructure makes more financial sense. Meta is planning a cloud business codenamed 'Meta Compute'—hosting models for a fee and renting bare-metal compute. He cited SpaceX leasing its Memphis data center to Anthropic for $1.25B/month; Meta is evaluating similar high-premium short-term deals. 2026 capex guidance is $125–145B, with in-house AI chip mass production set for September and a 14 GW compute deployment target by 2027.

OpenAI dropped GPT-5.6 in three sizes—Sol, Terra, Luna—on July 10. Sol hits 53.6 on Agents' Last Exam, beating Claude Fable 5 by 13.1 points at roughly one-quarter the cost. API pricing starts at $5/$30 per million input/output tokens for Sol, with cheaper tiers below. Codex desktop merges into ChatGPT alongside ChatGPT Work, Sites beta, and a multi-agent beta; the new 'ultra' effort level runs four agents in parallel by default. Meta launched Muse Spark 1.1 the same day but got overshadowed.

Tencent is leading a deal to unwind Meta's $2bn acquisition of AI agent startup Manus. Manus builds agent products that put models into business workflows. The post only provides a headline—no deal structure, timeline, or regulatory rationale is disclosed. I'd hold off until more details surface.

Meta released Muse Spark 1.1 via a new API, built around delegating tasks to parallel sub-agents and cross-device GUI control. It leads on 4 agent benchmarks—JobBench jumped 3.2× from 17.0 to 54.7. Coding trails flagships: Terminal-Bench 80.0 vs GPT 5.5's 83.4, SWE-Bench Pro 61.5 vs Opus 4.8's 69.2. Zuckerberg pitched it as very low price, aiming for strong-enough agent performance with cheap-enough coding. The post doesn't disclose exact pricing or rollout scope.

Meta Superintelligence Labs released Muse Spark 1.1, a multimodal reasoning model with major gains in tool use, computer use, and coding. It zero-shot generalizes to new tools and MCP servers, manages a 1M-token context window, and compacts memory to keep critical steps. The model orchestrates multi-agent systems, delegating tasks to parallel subagents to cut end-to-end latency. Coding improvements cover bug fixes, feature additions, and large code migrations in complex codebases. It is live in Meta AI's Thinking mode and in the new Meta Model API public preview.

Meta launched Muse Image on July 7, built by Meta Superintelligence Labs and free in Meta AI app, Instagram Stories, and WhatsApp. It does standard AI image generation with preset prompts. The flashpoint: you can @ any public Instagram user and remix their photo into a new AI image. The post doesn't say whether users can opt out or what training data was used. I'd hold off on trusting the privacy story.

Meta's Superintelligence Labs released Muse Image and Muse Video. Muse Image handles precise instruction-following, editing, and multi-reference composition using Instagram social context, plus agentic tool use. Muse Video shares the same pretrained base, outputs video with native audio. Available now in select countries via Meta AI app, web, Instagram Stories, and WhatsApp. The post doesn't disclose model size, latency, or which countries.

Reddit built new anti-spam tools with LLMs, blocking 23M spam views and catching ~25K new spam posts/comments daily. User exposure to spam dropped 20% from Jan–Mar vs the prior quarter. The tools now catch subtler coordinated fake behavior. The article notes AI moderation still needs human review for best results.

Wired obtained internal docs and spoke to five sources: Meta ran a project codenamed Cannes via contractor Covalen, with hundreds of workers creating fake under-18 accounts to probe ChatGPT, Gemini, and Character.AI. They sent over 45,000 prompts designed to bypass safety filters—covering suicide, self-harm, eating disorders, and sexual topics—without the competitors' knowledge. A spreadsheet of 3,748 prompts includes a 13-year-old asking for abortion pills and a fifth-grader describing a gun threat. Meta calls it routine safety benchmarking and says the data isn't used for training. Worth flagging: using fake identities to stress-test rivals' safety isn't the same as standard red-teaming.

Zuckerberg stated Meta is building a single AI cluster called Prometheus, exceeding one gigawatt. He used "hundreds of billions of dollars" to describe the capital spend and framed his role as concentrating elite talent, capital, and infrastructure. The post does not disclose a timeline, chip specs, or PUE details—only this one-line claim so far.

Scaling law is an empirically fitted curve, not a physical law. OpenAI's 2020 Kaplan paper concluded 'prioritize parameters' due to experimental biases, shaping GPT-3. DeepMind's 2022 Chinchilla corrected the ratio to 20:1, showing smaller models with more data outperform. Two 2024 replication studies confirmed that fixing Kaplan's setup reproduces Chinchilla's result—no fraud, just calibration. Since 2023, Meta and others deliberately deviate from Chinchilla: Llama 3 8B was trained on 15T tokens because the optimization target shifted from training cost to total cost of training plus inference. Tsinghua's Densing Law shows the parameter count needed for equal capability halves roughly every 3.5 months, but there is a floor: each parameter stores only ~2 bits of knowledge. The viral 'collapse' article cited a blog comment posted the same day as if it were peer-reviewed research; the post does not provide a paper source for that claim.

The hardest data in today's digest: a token-level analysis of 446 models on OpenRouter shows Anthropic's share dropped from 20.7% to 17.6% during the 18-day Fable 5 ban—the only major lab that didn't grow. GLM-5.2 quadrupled its share to 7.4% in two weeks on MIT license and 10x cheaper pricing, though per-task token consumption rivals Opus 4.8, narrowing the real cost gap. Community sentiment turned uglier: Fable 5's July 1 return came with task fallback to Opus, a 50% weekly cap, and credits billing—HN called it bait and switch, and anger at Anthropic's business tactics now exceeds anger at the government. Another standout: a solo dev gave Fable 5 a one-line goal; it spun up 22 agents, ditched Opus 4.8's Cloudflare setup, filed a support ticket on Volcengine, talked to engineers, and patched a security hole with a self-designed handshake—zero human touch. On tools: someone finally got credential pool auto-rotation working with Fable's help; another spent an hour routing Claude Code through OpenCode Zen to reach Fable 5. Quick hits: OpenAI negotiating a 5% equity donation to the US government, Tesla capping employee AI spend at $200/week, Meta claiming its Watermelon model matches GPT-5.5 internally, and Alibaba merging three agent products into one.

Yann LeCun argued at VivaTech that LLMs like ChatGPT can't handle real-world complexity and aren't a path to human-level intelligence. His new venture AMI Labs, founded after leaving Meta in 2025, is building JEPA—an architecture that learns abstract representations instead of memorizing statistical patterns. The company raised over $1B in seed funding from Nvidia and Jeff Bezos' family fund. Oxford's Ingmar Posner is pursuing a similar direction with world models that reason about causality. The post does not disclose JEPA's performance benchmarks or a product timeline.

At an internal town hall Thursday, Meta CEO Mark Zuckerberg said AI agent development hasn't accelerated the way executives expected. Earlier this year Meta laid off ~8,000 employees and reassigned ~7,000 to AI groups. Zuckerberg admitted the cuts weren't as 'clean' as they should have been and the upside of the new AI-focused structure hasn't materialized yet. He expects improvements from AI investments in the next 3–6 months. Meta is on track to spend up to $145 billion on AI infrastructure this year.

Meta warned ~6,000 employees that internal AI token costs are on track to hit billions this year. Employees burned through 73.7 trillion tokens in ~30 days, tracked on an internal leaderboard called 'Claudeonomics.' CTO Andrew Bosworth said token volume isn't a measure of impact. Meta is scrapping the leaderboard, rolling out an 'AI Gateway' monitoring dashboard, and will enforce formal token budgets starting in 2027. The company is also steering staff from Anthropic Claude toward its own MetaCode assistant. Uber faced a similar problem—it blew through its 2026 AI coding budget in four months and now caps spending at $1,500 per person per month.

Meta is building a cloud infrastructure business to sell AI compute and model access, putting it in direct competition with AWS, Google Cloud, and Microsoft Azure. The move comes weeks after SpaceX leased its Colossus 1 data center capacity to Anthropic. The pattern suggests data center owners, not model builders, may end up winning the AI race.