daily brief: 2026-05-18
musk's openai lawsuit dismissed, new tools for local ai and ocr, and research on hidden bias in language models.
ailatest summaries
plain 200-word summaries of important ai and data science news, updated every few hours.
paddleocr 3.5 adds transformers backend
paddleocr 3.5 lets developers run ocr and document parsing models with a hugging face transformers backend, reducing integration friction for rag and document ai workflows.
aimusk loses openai lawsuit on statute of limitations
a california jury found elon musk's claims against openai and sam altman were filed too late, removing a major legal threat before openai's reported ipo.
aithe hidden skill gap beyond sql and python
data jobs now demand data modeling, performance optimization, infrastructure awareness, and practical ai skills beyond basic sql and python.
aipytorch 2.11 fixes aarch64 gpu wheel installs
pytorch 2.11 now publishes cuda-enabled wheels for aarch64 linux on pypi, removing the need for custom indexes and workarounds when deploying on nvidia grace hopper and grace blackwell systems.
aipytorch executorch mlx delegate for apple silicon gpus
a new executorch backend enables gpu-accelerated inference on apple silicon macs using apple's mlx framework, with broad model and quantization support.
ailatent bias in llms can flip decisions
instruction-tuned language models show fair outputs but retain biased internal representations that can reverse decisions when activated.
aifine-tune cosmos predict 2.5 with lora for robot videos
a guide to parameter-efficient fine-tuning of nvidia cosmos predict 2.5 using lora and dora for generating synthetic robot manipulation videos.
aialexa+ can now generate podcast episodes on demand
amazon's alexa+ gets a new feature that creates ai-generated podcast episodes from any topic, with customizable length and tone.
aiteamtr fine-tunes multi-agent llms with trust regions
a trust-region method for fine-tuning multi-agent llm teams avoids compounding errors from stale rollouts, outperforming baselines by 7.1%.
aiopen agent leaderboard compares full ai systems
a new open benchmark evaluates complete agent systems, not just models, across six diverse tasks to measure generality, quality, and cost.
aifive practical uses for local language models
a technical writer shares five real-world tasks done with local llms, from private document search to offline assistants and code review, showing local models can be better than cloud for privacy and control.
ailetinar builds thin optics for ai glasses
south korean startup letinar develops pintilt lens tech to make ai smart glasses thinner, lighter, and more power-efficient.
aideepslide: ai for full presentation delivery
deepslide is a multi-agent system that helps prepare entire presentations, from planning and slide creation to rehearsal support.
aioracle price map learning for contextual pricing
a new method learns the optimal price as a smooth function of a scalar index, improving revenue in semiparametric dynamic pricing.
aitheory of mind gains don't always help human-ai chats
improving llm theory of mind on static tests often fails to boost real interactive task performance, a new study finds.
aistate-constrained dispatch cuts multi-agent alignment tax
a new framework uses state machines and online rlhf to enforce business rules in multi-agent systems, improving accuracy over gpt-4o on a recruitment benchmark.
aialpha-tcav fixes statistical flaws in concept-based explainability
a new framework replaces the unstable indicator function in tcav with a smooth parameterized function, reducing variance and unifying existing methods.
ai