Potentially, but needs more testing. Most other models are trained on a claude code-like harness, whereas codex is different enough to justify modifications. You can change a gate line in hackable install to test. If I enable it for all and it does not help, I get shit here
RESEARCH
-
AISLE vs Mythos: LLM Vulnerability Scanning Benchmark Debate
By
–
The moment you've read the analysis done by AISLE, you'll realize that the criticism made here has no basis. The reason is that AISLE hasn't actually reproduced the task that Mythos has solved (letting the LLM scan entire projects until it finds vulnerabilities without human
-
Anthropic Refuses to Release Its Too Powerful Mythos AI Model
By
–
Anthropic has created an AI model so powerful that it refuses to release it. Mythos discovered thousands of critical zero-day vulnerabilities in the world's most used software within days. Amazon, Apple, Microsoft, Google are already testing it. An Anthropic engineer: "I found more bugs in two weeks than in the rest of my entire life." [Translated from EN to English]
→ View original post on X — @alex_tsico, 2026-04-12 11:39 UTC
-
Human intelligence versus AI capability debate continues
By
–
And some say all humans are way smarter than the most capable AI pic.twitter.com/2T21G3VtdR
— AI (@DeepLearn007) 12 avril 2026And some say all humans are way smarter than the most capable AI
-
Atlas Robot Learns Human-Inspired Movements Through Reinforcement Learning
By
–
Reinforcement-Learned Atlas Performs Moves Inspired by Human Mocap and Animation
— Ronald van Loon (@Ronald_vanLoon) 12 avril 2026
via @ZappyZappy7
#Robotics #Engineering #ArtificialIntelligence #Innovation #Technology pic.twitter.com/4zXXce8uXwReinforcement-Learned Atlas Performs Moves Inspired by Human Mocap and Animation via @ZappyZappy7 #Robotics #Engineering #ArtificialIntelligence #Innovation #Technology
→ View original post on X — @ronald_vanloon, 2026-04-12 10:18 UTC
-
Claude Mythos Linked to ByteDance’s Looped LLM Research
By
–
This might be the most important observation about Claude Mythos that nobody is picking up on. Chris is connecting Mythos to ByteDance’s “Scaling Latent Reasoning via Looped Language Models” paper. The core idea: instead of thinking out loud with chain-of-thought text, a looped
-
Oscar: The Living Robot That Can Heal Itself
By
–
Meet Oscar: The Living #Robot That Can Heal Itself
— Ronald van Loon (@Ronald_vanLoon) 12 avril 2026
by @pascal_bornet
#Robotics #AI #Biotech #Innovation #FutureOfTechnology pic.twitter.com/BLrPrV8X6PMeet Oscar: The Living #Robot That Can Heal Itself by @pascal_bornet #Robotics #AI #Biotech #Innovation #FutureOfTechnology
→ View original post on X — @ronald_vanloon, 2026-04-12 09:19 UTC
-
Memory Sparse Attention GitHub Repository Reaches 3000 Stars
By
–
讲真，我们没有想到 Memory Sparse Attention 的 GitHub Repo 会这么快到 3000 星。更没有想到大家的关注度这么高。也许真的印证了那句话吧，「能带走的才是记忆，带不走的则是过去」。
— 艾略特 (@elliotchen100) 12 avril 2026
起初，MSA 只是一篇论文。一个支持 1 亿 Token 上下文的端到端方案，大胆，也优雅。
核心就两个点：一个是 Dual… pic.twitter.com/DUBidTcl8rTo be honest, we didn't expect the GitHub Repo for Memory Sparse Attention to hit 3000 stars so quickly. Even less did we expect such high attention from everyone. Maybe it really proves that saying: "What you can take with you is memory; what you can't take is the past." At the
-
Hassabis vs LeCun: Major AI Researchers Clash on LLM Future
By
–
The CEO of Google DeepMind just went on record saying he disagrees with one of the most respected AI researchers in the world.
— Milk Road AI (@MilkRoadAI) 12 avril 2026
Demis Hassabis, the man behind AlphaFold, AlphaGo, and Google's entire AI operation publicly pushed back against Yann LeCun's claim that large language… pic.twitter.com/qmrLXNEqXUThe CEO of Google DeepMind just went on record saying he disagrees with one of the most respected AI researchers in the world. Demis Hassabis, the man behind AlphaFold, AlphaGo, and Google's entire AI operation publicly pushed back against Yann LeCun's claim that large language models are a dead end for artificial intelligence. LeCun, who left Meta earlier this year to start his own AI lab, has been saying for years that LLMs cannot reason, cannot plan, and will never get us to human-level intelligence. Hassabis disagrees, and he said so directly. His position is that scaling laws are still working, foundation models are still getting more capable, and whatever AGI ends up looking like, LLMs will be a central part of it, not something that gets replaced. He does say there is roughly a 50/50 chance that one or two additional breakthroughs will be needed beyond scaling alone, things like better memory, long-term planning, and world models. But the core disagreement with LeCun is clear, Hassabis believes the current architecture is sound and the current path leads somewhere real. Two Nobel-recognized researchers, two founding figures of modern AI, now publicly on opposite sides of the most important technical question in the industry.
→ View original post on X — @ceobillionaire, 2026-04-12 09:02 UTC
-
UnMarker: A Quiet Breakthrough Beyond SynthID Debate
By
–
But that’s not the FULL story. Because while everyone’s debating whether one researcher “jailbroken” SynthID, something much bigger happened quietly in the background. A team at the University of Waterloo built a tool called UnMarker. Published at IEEE Security and Privacy.