AI Coding Live News

At 18:58 UTC on November 9 (2:58 a.m. on November 10 SGT), the official Claude account released Claude Fable 5.2 and Claude Mythos 5.2. They are the same underlying model: Fable ships with cybersecurity and biology safeguards and is open to all paid plans, the Claude Platform and the major clouds, while Mythos drops some of those safeguards and is offered only to vetted organizations through the Cyber and Life Sciences Verification Programs. Pricing falls from Fable 5.1's $10/$50 to $8/$40 per million input/output tokens, output is up to 30% faster, and an optional fast mode generates up to 2.5x faster for $16/$80. By Anthropic's numbers, Fable 5.2 at Xhigh effort scores 49.1% on FrontierCode 2.0, ahead of Opus 5.5 (45.5%) and GPT-6 Astra (42.8%); at High effort it matches Opus 5.5's best for about half the cost per task. The API model name is claude-fable-5-2.

Zoom

On November 5 Stripe made agent access to Stripe analytics available to all users via the Stripe MCP or the Queries and Reports APIs directly. Compatible agents such as Claude and Cursor can answer natural-language questions about MRR, churn, fraud and reconciliation. The Reports API ships a modeled schema of 300+ tables with built-in guidance; Queries API is free through January 31, 2027 then usage-based, while Reports API stays free. Every analysis exposes the underlying SQL for verification.

Zoom

In an Oct 26 blog post, OpenAI said its internal models produced three new results: a proof of Sogge's local smoothing conjecture in every dimension (which implies the Kakeya and Stein restriction conjectures), a Lean-verified proof that the Exponential Time Hypothesis (ETH) is false, and a rigorous existence proof for four-dimensional QCD and related quantum field theories including pure Yang-Mills. The math and CS papers and Lean proofs are in the openai/math GitHub repo; the physics artifacts are being held back while an external group led by David Gross writes them up. All of this is OpenAI's own claim and still awaits independent verification.

Zoom

GitHub Next officially released PDD Kit (Property-Driven Development Kit), establishing a formal verification toolkit tailored for autonomous coding agents. Addressing the pervasive reliability gap where agent-generated code passes isolated unit tests but crashes on subtle edge cases, PDD Kit combines property-driven specifications with automated formal solvers. Instead of testing arbitrary sample inputs, PDD Kit mathematically proves that agent code adheres to rigorous invariant specifications across every possible input, preventing regressions and elevating coding agents from probabilistic generation to provable correctness.

Zoom

OpenAI announced a critical coordinated vulnerability disclosure in the core PHP runtime discovered by cybersecurity firm Calif utilizing GPT-5.6 Sol, alongside independent verification by Wiz. Because PHP powers roughly 70% of all websites globally, the vulnerability represented an existential risk to internet infrastructure. Following coordinated patching, OpenAI committed $1B in subsidized Daybreak security access and open-sourced openai/php2shell-checker, an automated Codex inspection tool enabling site administrators worldwide to audit their deployments.

Zoom

Scale AI Research introduced Sauron (arXiv:2610.17724), an open-source evidence-first agentic verification framework. Overcoming the severe failure modes of standard LLM-as-a-judge approaches—which hallucinate justifications and guess outcomes over long agent trajectories—Sauron enforces bounded, adaptive interrogation of agent traces with cryptographic-grade execution receipts. Evaluated on the PA100 benchmark across eight backbones, it delivers 0.94 precision and 0.97 recall at roughly 15% the cost of state-comparison baseline Archipelago, completing 10-episode grading 4.2x faster.

Zoom

Researchers from Amazon Web Services (AWS) and UIUC unveiled Lean4AgenticRL (arXiv:2610.22448). Tackling the limitations of conventional agentic reinforcement learning relying on sparse, binary outcome rewards, the framework grounds multi-level process rewards within a formal Lean4 verification schema. Utilizing workflow-stage-normalized GRPO, the resulting LAR-27B model reaches 77.3% on TerminalBench-2.1, outperforming base models by 9.89% and surpassing comparable open models by 5.07%.

Zoom

Three new studies tackle verification, evaluation, and RL alignment for rule-governed generative world models: OmniVerifier-C1 (arXiv:2610.22355, 32 upvotes) uses Decoupled RL to train a decomposed graded visual verifier that reaches 0.684 on ViVerBench and boosts FLUX.1-dev GenEval from 0.68 to 0.79 while enabling Maze and Sudoku world modeling; Meituan's WorldCast benchmark (arXiv:2610.20731, 30 upvotes) evaluates 12 models across 320 multi-step world state prediction cases, finding GPT-Image-2.5 System at 72.8 vs. 42.8 for Nano Banana Pro and 20.1 for the top open model; and NUS's Routing Policy Optimization (RPO, arXiv:2610.22282, 23 upvotes) turns attention routing into an internal exploration action to train video/image diffusion RL 3.1x–4.2x faster than Flow-GRPO and DanceGRPO.

Zoom

Three new frameworks advance web and multimodal agent post-training: Salesforce AI Research's CLIFT (arXiv:2610.06829) certifies URL-conditional self-verification questions for dense RL rewards and judge-free Conformal Trajectory Selection (CTS), propelling Gemma-4 31B to a 74.6% SOTA on WebArena Infinity (beating Gemini 3 Flash at 70.1%) and lifting GPT-5.5 to 53.7% on VisualWebArena. Meanwhile, MMSearch-OPD (arXiv:2610.22240) boosts 9B multimodal deep search by +15.3 points to 60.2%, and Northeastern University's SCOPED-8B (arXiv:2610.22217) uses skill-decoupled on-policy expert distillation to reach 73.13% across 8 benchmarks—surpassing Qwen3-VL-32B (72.01%) and Claude Opus 4.6 (72.86%).

Zoom

Tackling step-level verification in mathematical reasoning and in-context copying bottlenecks in Transformers, NYU/UCLA/Stanford researchers introduce LeanPRM (arXiv:2610.17697)—using the Lean 4 theorem prover to automatically label 152K steps across 10,634 natural-language proofs with zero human annotation, boosting Qwen3-4B step accuracy on LeanPRM-Bench from 48.2% to 74.7% (vs. 60.3% for PRM800K). Meanwhile, Cornell University's Copy-Paste Language Models (CPLMs, arXiv:2610.19247) add a lightweight Copy-Paste head (~3% compute overhead) that relieves the LM head gradient bottleneck, boosting LAMBADA accuracy by +2.5-4.0 points and allowing a w=128 sliding-window CPLM to match full-attention perplexity.

Zoom