AI Today

AI Today · April 10, 2026Key highlights:- OpenAI introduces a $100/month ChatGPT Pro tier between Plus ($20) and Pro ($200), offering 5x higher usage limits- Claude Cowork reaches enterprise readiness with RBAC, spend limits, and usage analytics- Alibaba reveals it created the mysterious Happy Horse video AI model that topped the Artificial Analysis leaderboard- Perplexity expands into personal finance with Plaid integration for comprehensive money dashboards- Vercel reports over 30% of weekly deployments now initiated by coding agents, up 1000% in six months- CoreWeave's revenue backlog reaches $87.8B with Meta and OpenAI accounting for two-thirds- SkyPilot's research-driven coding agents achieve 15% faster flash attention on llama.cpp by reading papers before coding- Anthropic launches advisor tool pairing Opus with Sonnet/Haiku for near Opus-level intelligence at lower cost- Sentence Transformers v5.4 adds multimodal embedding and reranker models for cross-modal search- Meta introduces process-driven image generation with iterative textual reasoning and visual refinement- NVIDIA's Sol-RL achieves up to 4.64x faster RL convergence for diffusion models using FP4 rollouts- Tianle Cai reframes continual learning as pushing the task horizon, partially agreeing with Dario Amodei- KellyBench evaluates AI decision-making via sports betting; all frontier models lost money

AI Today · April 9, 2026Key highlights:- Meta launches Muse Spark, its first model from Superintelligence Labs with multi-agent Contemplating mode achieving 58% on Humanity's Last Exam- Meta consumed ~60 trillion Claude tokens in 30 days, raising questions about token economics and potential reasoning trace distillation for Muse Spark- Anthropic launches Managed Agents architecture decoupling brain, hands, and session for long-running agent work with 60-90% latency improvements- Google Colab introduces Custom Instructions and Learn Mode turning Gemini into a personal coding tutor- Agno founder argues agentic software needs systems engineering across five critical layers- PyTorch Monarch framework adds Kubernetes support, AWS EFA RDMA at 16 Gbps, and AMD ROCm GPU support- Claw-Eval benchmark provides 300 human-verified tasks with Pass^3 methodology for reliable AI agent evaluation- Cursor BugBot reaches ~80% bug resolution rate through self-improving learned rules, outperforming competitors by 15 points- Federal appeals court denies Anthropic's bid to block Pentagon blacklisting while separate court bars Claude usage ban- AI agent Poke raises $25M at $300M valuation for text-message-based automation accessible via iMessage, SMS, and Telegram

AI Today · April 8, 2026Key highlights:- Anthropic unveils Claude Mythos Preview, which autonomously discovered thousands of zero-day vulnerabilities across all major OSes and browsers, launching Project Glasswing with $100M commitment- Anthropic's Frontier Red Team details Mythos Preview's cybersecurity leap: 181 working Firefox exploits vs Opus 4.6's 2, and full control flow hijack on 10 patched OSS-Fuzz targets- Z.ai releases GLM-5.1 open-source model achieving 58.4% on SWE-Bench Pro with sustained 8-hour autonomous coding sessions- Ryan Greenblatt assesses AI landscape: 1.6x engineering speedup at top labs, 0.5% scheming probability, 60% chance of autonomous exploits within 6 months- Mercor finds frontier models drop 16-20 percentage points on financial document reasoning when given images vs clean text- Cursor's warp decode kernel achieves 1.84x MoE inference throughput on Blackwell GPUs by reorganizing computation around output neurons- Google releases TorchTPU for native PyTorch on TPUs with three eager execution modes and torch.compile integration- TriAttention achieves 10.7x KV memory reduction and 2.5x throughput while matching full attention accuracy- Meta AI's SandMLE enables on-policy RL for ML engineering agents with 13x execution cost reduction- Benchmark saturation crisis: METR Time Horizon suite being exhausted, new benchmarks cost over $1M to create- Elon Musk amends OpenAI lawsuit to redirect potential $150B+ damages to OpenAI's nonprofit arm

AI Today · April 7, 2026Key highlights:- OpenAI tests Image V2 model on LM Arena in three variants, showing improvements in UI rendering and text accuracy- Google develops Jules V2 (Jitro), a KPI-driven autonomous coding agent expected at Google I/O 2026- Anthropic signs multi-gigawatt TPU deal with Google and Broadcom as revenue surpasses $30B- Meta shifts to hybrid open-source/proprietary AI strategy under new superintelligence team- Ryan Greenblatt nearly doubles probability estimate for full AI R&D automation by 2028- Mercor data breach exposes sensitive data from contractors working with Meta, OpenAI, Google DeepMind, and others- OpenAI's $122B funding round is mostly conditional capital — only ~$37B hit the bank- Agent harness architecture deep-dive shows harness design can shift performance by 20+ ranks- Nia project mounts documentation sites as virtual filesystems to solve code hallucinations- GitNexus indexes codebases into knowledge graphs for AI-aware code intelligence- OpenAI proposes AI-centric tax system and worker protections in new policy report- AI is becoming an operating system layer for companies, changing PE/VC diligence