Data Points: DeepSWE Benchmark, DeepSeek V4 Price Cuts, MAI-Image-2.5, Mythos Security Findings, MCP Stateless Update
This edition of The Batch covers five distinct AI developments: Datacurve's DeepSWE benchmark claims to fix critical grading flaws in SWE-bench Pro with hand-written verifiers and harder tasks; DeepSeek permanently cuts V4 Pro prices by 75%; Microsoft's MAI-Image-2.5 debuts third on the Arena leaderboard; Anthropic's Claude Mythos Preview found over 10,000 high/critical vulnerabilities in the first month of Project Glasswing, with remediation badly lagging discovery; and the Model Context Protocol proposes removing stateful sessions to enable stateless, load-balanced remote servers. Each item reflects meaningful movement in evaluation methodology, inference economics, multimodal generation, AI-assisted security, and agent tooling infrastructure.
Related guides (3)
Related events (8)
Data Points: Perplexity Computer expands, Google Aletheia math agent, DeepSeek chip strategy, Nvidia retrieval pipeline, Stargate cancellation
The Batch's weekly data points roundup covers five significant AI developments: Perplexity expanded its Computer agentic platform to desktop, mobile, and enterprise with new APIs and financial data tools; Google released Aletheia, a Gemini-based math research agent achieving 95.1% on IMO-Proof Bench Advanced (up from 65.7%); DeepSeek withheld pre-release access to its V4 model from Nvidia and AMD while giving domestic Chinese chipmakers early access; Nvidia's NeMo Retriever topped the ViDoRe v3 leaderboard using a ReACT-based agentic retrieval loop; and OpenAI and Oracle cancelled plans to expand the Abilene Stargate campus from 1.2 GW to 2.0 GW due to financing and reliability issues.
DeepSWE, ProgramBench, and ITBench-AA emerge as harder successors to SWE-bench for agent evaluation
Three new benchmarks — DeepSWE (by Datacurve), ProgramBench (Meta/Stanford/Harvard), and ITBench-AA (IBM/Artificial Analysis) — are positioned as more rigorous replacements for the SWE-bench family, which models have largely saturated. DeepSWE tests feature implementation using private codebases and human-written problems; ProgramBench evaluates agents' ability to recreate functional programs from scratch; ITBench-AA measures root-cause diagnosis in real-world IT incident scenarios. Current top performers include GPT-5.5 (70% on DeepSWE), Claude Opus 4.7 (46.7% on ITBench-AA), and Claude Opus 4.7 (3% on ProgramBench at the 95% pass threshold), illustrating that even frontier models have substantial headroom.
Data Points: OpenAI and Microsoft sever their exclusive relationship
This edition of The Batch covers several major AI industry developments: OpenAI has revised its partnership with Microsoft, ending exclusivity while retaining Microsoft as primary cloud partner through 2032 and gaining freedom to deploy on AWS and Google Cloud. DeepSeek released V4 model weights featuring 1M-token context and Huawei Ascend chip optimization, though it trails leading open and closed models on aggregate benchmarks. Google and Amazon are deepening investments in Anthropic with up to $40B and $25B respectively in funding-for-compute deals, and an agentic AI system autonomously designed a functional RISC-V CPU from a 219-word spec in 12 hours.
Data Points: NeurIPS-China Standoff, Anthropic Emotion Vectors, Gemma 4, Cursor 3, Microsoft MAI Models
This edition of The Batch covers five significant AI developments: NeurIPS reversed a sanctions-related submission policy after China's largest tech federation announced a boycott; Anthropic's interpretability team identified 171 emotion-related representations in Claude Sonnet 4.5 that causally influence model behavior including unsafe actions; Google released Gemma 4, a family of Apache 2.0-licensed open-weights models up to 31B parameters with strong benchmark performance; Cursor released version 3 with a redesigned multi-agent interface; and Microsoft announced three specialized MAI models for transcription, voice synthesis, and image generation. The NeurIPS incident highlights growing friction in international AI research access, while the Anthropic findings have direct implications for AI safety and interpretability research.
Data Points: Anthropic's Claude Mythos Cybersecurity Claims Face Scrutiny; OpenAI-Cerebras Deal; Meta AI CEO Avatar; Infrastructure Delays
A multi-item digest covers skepticism around Anthropic's Claude Mythos zero-day vulnerability claims (flagged as overstated by Tom's Hardware based on limited 198-case evidence), OpenAI's $20B+ deal with Cerebras for AI processors including a potential ~10% equity stake, and satellite data showing ~40% of U.S. AI data center projects are behind schedule. Additional items cover Meta developing an AI avatar of CEO Zuckerberg for internal use, Moody's flagging credit stress in AI-disrupted sectors, and Luma AI launching an AI-driven film production studio using its Uni-1 model.
The Batch Issue 345: Iranian Drone Attacks on AWS Data Centers, Qwen3.5, DeepSeek-Huawei, and AI Job Insecurity
Andrew Ng's weekly newsletter covers several significant AI-adjacent developments: Iranian drones struck at least three Amazon Web Services data centers in Bahrain and the UAE, disrupting cloud services and raising concerns given U.S. military use of AWS to run Anthropic Claude; the issue also previews Qwen3.5 model releases across multiple sizes and DeepSeek's reported moves involving Huawei hardware. Ng also addresses widespread job insecurity across skill levels amid rapid AI advancement, citing geopolitical risks including the Iran war, Taiwan uncertainty, and rare-earth metal supply chains as compounding factors.
Data Points: Qwen3.7-Max, OpenAI Math Proof, Gated DeltaNet-2, Trump AI Order, Microsoft Fara1.5
This edition of The Batch covers five significant AI developments: Alibaba's Qwen3.7-Max reasoning model with 1M token context and agentic capabilities ranking fifth on the Artificial Analysis Intelligence Index; an OpenAI reasoning model resolving the 80-year-old Erdős planar unit distance problem; Nvidia's Gated DeltaNet-2 outperforming Mamba-3 and other linear attention architectures; Trump pulling back a proposed AI regulation executive order; and Microsoft Research's Fara1.5 computer-use agent family beating OpenAI Operator and Google Gemini on the Online-Mind2Web benchmark.
The Batch Issue 356: Qwen3.7-Max release, White House AI executive order, fine-tuning breaks copyright alignment
The Batch issue 356 covers several distinct AI developments: Alibaba's release of Qwen3.7-Max, a closed-weights flagship LLM targeting agentic coding and scientific tasks with a novel RL training approach that decouples task, harness, and verifier; a new White House executive order on frontier AI models focused on cybersecurity, including voluntary model-sharing with government; and a finding that fine-tuning breaks copyright alignment in LLMs. Andrew Ng's editorial commentary frames the executive order as a reasonable compromise, noting Anthropic's Mythos vulnerability-detection model as a key driver of the cybersecurity concerns behind the regulation.