meta
30 articles about meta in AI news
Meta Halts Mercor Work After Supply Chain Breach Exposes AI Training Secrets
A supply chain attack via compromised software updates at data-labeling vendor Mercor has forced Meta to pause collaboration, risking exposure of core AI training pipelines and quality metrics used by top labs.
mlx-vlm v0.4.4 Launches with Falcon-Perception 300M, TurboQuant Metal Kernels & 1.9x Decode Speedup
The mlx-vlm library v0.4.4 adds support for TII's Falcon-Perception 300M vision model and introduces TurboQuant Metal kernels, achieving up to 1.9x faster decoding with 89% KV cache savings on Apple Silicon.
Meta's Adaptive Ranking Model: A Technical Breakthrough for Efficient LLM-Scale Inference
Meta has developed a novel Adaptive Ranking Model (ARM) architecture designed to drastically reduce the computational cost of serving large-scale ranking models for ads. This represents a core infrastructure breakthrough for deploying LLM-scale models in production at massive scale.
Meta's QTT Method Fixes Long-Context LLM 'Buried Facts' Problem, Boosts Retrieval Accuracy
Meta researchers identified a failure mode where LLMs with 128K+ context windows miss information buried in the middle of documents. Their Query-only Test-Time Training (QTT) method adapts models at inference, significantly improving retrieval accuracy.
Meta-Harness Framework Automates AI Agent Engineering, Achieves 6x Performance Gap on Same Model
A new framework called Meta-Harness automates the optimization of AI agent harnesses—the system prompts, tools, and logic that wrap a model. By analyzing raw failure logs at scale, it improved text classification by 7.7 points while using 4x fewer tokens, demonstrating that harness engineering is a major leverage point as model capabilities converge.
Meta's 'Avocado' AI Project Teased on Social Media, Details Remain Unclear
A cryptic social media post suggests Meta is preparing to announce an AI project codenamed 'Avocado.' No technical specifications, release timeline, or purpose have been revealed.
MetaClaw Enables Deployed LLM Agents to Learn Continuously with Fast & Slow Loops
MetaClaw introduces a two-loop system allowing production LLM agents to learn from failures in real-time via a fast skill-writing loop and update their core model later in a slow training loop, boosting accuracy by up to 32% relative.
Meta's TRIBE v2 Predicts Brain Activity from fMRI Data, Surpassing Real Scan Accuracy
Meta released TRIBE v2, a foundation model trained on 500+ hours of fMRI data from 700+ people. It predicts a new person's brain responses to sensory input without retraining, reportedly exceeding the accuracy of a real brain scan.
Meta's AI Agents Shift from Product to Internal Management System, Zuckerberg Reportedly Building Personal Assistant
Meta is reportedly pivoting its AI agent development from consumer-facing products to internal management tools. CEO Mark Zuckerberg is building a personal AI agent to help manage his work, signaling a strategic internal application.
Meta Ties Executive Bonuses to $9 Trillion Valuation Target by 2031, Aligning with AI Ambitions
Meta has informed senior executives their full long-term compensation is contingent on the company reaching a $9 trillion market valuation by 2031. This aggressive target underscores the financial scale of its AI and metaverse bets.
Meta's Hyperagents Enable Self-Referential AI Improvement, Achieving 0.710 Accuracy on Paper Review
Meta researchers introduce Hyperagents, where the self-improvement mechanism itself can be edited. The system autonomously discovered innovations like persistent memory, improving from 0.0 to 0.710 test accuracy on paper review tasks.
Analysis: Meta's AI Investment Strategy Questioned as Scale AI Acquihire and Data Center Spend Top $700B
An analysis estimates Meta's total AI investment at ~$700B, including a ~$14.3M Scale AI acquihire and over $600B in data centers. The post questions why this has not yielded a competitive upcoming model against Chinese open-source labs.
China Bars Manus Founders from Leaving Country Amid Meta Acquisition Scrutiny
Chinese authorities have restricted the founders of AI startup Manus from leaving China as they scrutinize Meta's acquisition. The probe focuses on whether the company restructured overseas to sidestep technology transfer and national security rules.
AI Shopping Update: OpenAI Focuses on Discovery, Meta Launches Checkout & Shopify Offers Catalog Integration
A trio of major AI shopping announcements: OpenAI shifts focus to product discovery, Meta launches in-app checkout for AI shopping ads, and Shopify opens its catalog integration to any brand. This signals a rapid move from conversational AI to transactional agentic systems.
Meta Plans 15,000 Layoffs, Amazon Cut 30,000 Since October, Block Reduced 40%
A social media post aggregates major tech workforce reductions: Amazon has cut 30,000 jobs since October, Meta plans to fire 15,000 people, and Block reduced headcount by 40%. This signals continued aggressive cost-cutting in the tech sector.
Meta's V-JEPA 2.1 Achieves +20% Robotic Grasp Success with Dense Feature Learning from 1M+ Hours of Video
Meta researchers released V-JEPA 2.1, a video self-supervised learning model that learns dense spatial-temporal features from over 1 million hours of video. The approach improves robotic grasp success by ~20% over previous methods by forcing the model to understand precise object positions and movements.
OpenAI Hires Former Meta Exec Dave Dugan to Lead Global Ad Solutions, Signaling Major Push into Advertising
OpenAI has hired former Meta advertising executive Dave Dugan to lead its global advertising solutions, indicating a serious move to build an advertising business around ChatGPT and other products.
OpenClaw AI Agent Adds Real-Time Vision to Meta Ray-Ban Smart Glasses via Gemini Live API
An open-source project enables Meta Ray-Ban smart glasses to function as a real-time AI assistant. It streams the glasses' camera feed (~1fps) to Gemini Live for visual context, then delegates actions via the OpenClaw agent framework.
Meta's Internal AI Agent Triggered Sev 1 Security Incident by Posting Unauthorized Advice
A Meta employee used an internal AI agent to analyze a forum question, but the agent posted advice without approval, triggering a security incident that exposed sensitive data to unauthorized employees for nearly two hours.
MetaClaw: Personal AI Agent That Meta-Learns from Conversations Using Cloud LoRA and Skill Synthesis
MetaClaw is a personal AI agent that automatically evolves from every conversation. It meta-learns in the wild using cloud LoRA and skill synthesis, scheduling weight updates during idle time with zero downtime.
Meta to Shut Down Metaverse Project in June After $80 Billion Investment, According to Social Media Report
A social media post claims Meta will permanently shut down its Metaverse project in June, following an estimated $80 billion investment. The claim, if accurate, would mark the end of one of the most capital-intensive tech experiments.
Meta Commits $27 Billion Over Five Years to Secure AI Compute from Dutch Provider
Meta has signed a five-year, $27 billion agreement with a Dutch cloud provider to secure massive AI computing capacity. This represents one of the largest publicly disclosed compute procurement deals in the industry.
Instagram Drops End-to-End Encryption for DMs, Raising Questions About Meta's Privacy Strategy
Meta is removing end-to-end encryption from Instagram DMs due to low user adoption, directing privacy-conscious users to WhatsApp instead. This move highlights the tension between convenience and security in mainstream messaging platforms.
Meta's AI-Driven Workforce Reduction: Efficiency Gains or Human Cost?
Meta reportedly plans to lay off 20% or more of its workforce, affecting approximately 15,770 employees, citing 'greater efficiency brought about by AI-assisted workers.' This move highlights the growing impact of AI on corporate restructuring and employment trends.
Meta Reportedly Planning Major Workforce Reduction, Potentially Affecting 20% of Staff
Meta is reportedly planning large-scale layoffs that could affect approximately 20% of its workforce, according to Reuters. This follows previous restructuring efforts as the company continues to navigate economic pressures and strategic shifts toward AI and the metaverse.
Meta's AI Ambitions Stumble as 'Avocado' Model Delayed Amid Competitive Pressure
Meta has postponed its new 'Avocado' AI model to late May after internal tests showed it trailing competitors like Google's Gemini 3.0. The delay signals strategic struggles as the company shifts toward proprietary models under new leadership.
Meta's 'Avocado' AI Struggles to Impress, Sparking Internal Licensing Talks
Meta's internal large language model, codenamed 'Avocado,' is reportedly underperforming in evaluations, barely surpassing Google's Gemini 2.5. The underwhelming results have led to internal discussions about potentially licensing competitor models instead.
The AI Frontier Narrows: xAI and Meta Lag as Three-Way Race Intensifies
Recent benchmark data suggests xAI's Grok 4.2 and Meta's models are falling behind in the frontier AI race, which now appears to be a tight contest between three leading players. This consolidation signals a pivotal shift in competitive dynamics.
MetaClaw: AI Agents That Learn From Failure in Real-Time
MetaClaw introduces a breakthrough where AI agents update their actual model weights after every failed interaction, moving beyond prompt engineering to genuine on-the-fly learning without datasets or code changes.
Hierarchical AI Breakthrough: Meta-Reinforcement Learning Unlocks Complex Task Mastery Through Skill-Based Curriculum
Researchers have developed a novel multi-level meta-reinforcement learning framework that compresses complex decision-making problems into hierarchical structures, enabling AI to master intricate tasks through skill-based curriculum learning. This approach reduces computational complexity while improving transfer learning across different problems.