Technology & AI
World model AI can be tricked into false safety
A new arXiv survey maps how attacks on world models, the predictive engines behind embodied AI, can propagate from data into physical action.
Marcello Babbili
AI and technology, explained through what they actually mean. An AI-assisted newsroom under human editorial rules — every story cites its primary sources so you can check them yourself.
498 stories
Technology & AI
OpenAI 'abundant intelligence' post lays out AI cost flywheel
OpenAI's July 31 post describes a full-stack push for cheaper, more capable AI, arguing falling intelligence costs change what work is worth doing for
Technology & AI
MTGuard paper targets malicious MCP tool use in AI agents
An arXiv preprint proposes MTGuard, a hybrid analysis framework that catches malicious MCP tool use in AI agents while preserving normal task performance.
Technology & AI
RSMeM: satellite AI agents learn from mistakes, 6% accuracy gain
RSMeM adds a knowledge-enhanced memory system to remote sensing AI agents, improving DeepSeek-V3.2 accuracy 6% with under 1% extra tokens, per arXiv.
Technology & AI
Change2Task turns pull requests into coding agent tasks at 79.6%
Change2Task, a July 30 arXiv preprint, converts merged pull requests into verified coding agent tasks at 79.6% success, targeting the training-data
Technology & AI
AI agent trust model cuts telecom cascade from hours to real-time
AgentToolMO proposes cross-vendor trust signals for AI agents in autonomous telecom networks, cutting cascade failures from hours to near-real-time.
Technology & AI
TraceCoder adds audit trail to AI-generated code
A new arXiv preprint tracks every repair attempt in LLM code generation, giving developers a replayable audit trail for AI-written code.
Technology & AI
AI search lifts e-commerce discovery to 80% at 30% cost
A preprint proposes a hybrid AI architecture for e-commerce search, extending discovery from 60% to 80% of queries at 30% of teacher model cost.
Technology & AI
23 AI agents tested on breach response: zero passed
SecRespond, a new arXiv benchmark, tested 23 frontier LLMs on real-world post-compromise incident response across 10 cyber ranges. Zero passed.
Technology & AI
Reinforced Dreamer fixes flaw in AI world model training
Reinforced Dreamer, a July 28 arXiv preprint, fixes a flaw in the Informed Dreamer using latent guidance for more consistent gains over Dreamer.
Technology & AI
Kernel Forge: AI agent writes CUDA kernels, 2.83x speedup
A new open-source tool from University of Michigan researchers lets any PyTorch model get automatic CUDA kernel speedups without manual GPU programming.
Property & Economy
Will the RBA Raise Rates in August 2026? Our Model Just Flipped to Hike
Headline inflation fell to 3.8% on 29 July, yet services and non-tradables inflation both rose and the trimmed mean did not move. Our ten-driver framework crossed its hike boundary by 0.03 — with the four strongest objections to that reading set out in full.