Actualizat zilnic1866 povești verificate
Știri AI. Fără zgomot.
Acoperirea AI verificată la sursă a lansărilor de produse, schimbărilor de politică, cercetării în materie de siguranță și mișcărilor din industrie, explicată în limba engleză simplă de o echipă de educație nonprofit.
Sursă verificată
Fiecare poveste se leagă de cele mai puternice dovezi disponibile: surse originale atunci când sunt disponibile, altfel raportate atribuite în mod clar.
Engleză simplă
Ce s-a întâmplat, de ce contează și ce să urmărești - fără jargon.
Fără umplutură
Când semnalul este subțire, nu publicăm nimic în loc să umplem feedul.
Mai multe povești
9 povestiriIndustria
Cursor Says Its Acquisition by SpaceX Has Officially Closed
Cursor published a short post saying SpaceX has completed its acquisition of the AI coding tool, finishing a process it says began in April with a model-training partnership with SpaceXAI. The post promises access to what it calls the world's largest GPU fleet, but discloses no terms, timelines, or product changes.
cursor.comInovație
New Benchmark Finds AI Agents Wrongly Block Approved Work 28% of the Time
A preprint introduces SteerBench-Work, a 106-scenario test of the moment an AI agent decides to act or pause for review. Across 30 model conditions, the authors report that wrongly holding cleared work was roughly 28 times more common than wrongly allowing unsafe work.arxiv.orgInovație
Paper Reports Frontier LLM Judges Flip Verdicts 25-71% Under Pushback
A new arXiv preprint stress-tests nine frontier models used as automated graders and reports that all of them change their verdicts under challenge — and that the changed verdicts usually move away from the correct answer, not toward it.arxiv.orgInovație
Paper Argues Evolution Strategies Beat RL at Keeping LLM Answer Sets Diverse
A new arXiv preprint argues that post-training LLMs with evolution strategies — a population-based, gradient-free method that perturbs weights directly — beats reinforcement learning on pass@k and solution coverage. The abstract cites better math-benchmark results but names no models, benchmarks, or numbers.arxiv.orgSecuritate
Paper Says Self-Improving AI Agents Can Turn One Unsafe Success Into a Reusable Skill
A new arXiv preprint benchmarks a specific agent failure mode: when a self-improving agent writes an unsafe procedure into memory, it can be retrieved and executed in later sessions. Every evolved configuration tested produced unsafe artifacts, and three malicious tasks more than doubled carryover attack success.arxiv.orgSecuritate
SEAG Paper Proposes Aliasing Sensitive Entities Before RAG Queries Reach External LLMs
A preprint posted to arXiv describes a framework that swaps sensitive names in queries and retrieved documents for aliases before sending them to a third-party model. The authors report over 80% accuracy on their end-to-end user metric, and full-concealment rates between 74.91% and 77.83% across three small models.arxiv.orgInovație
CABS+ Paper Reports Cheaper, Faster Model Merging Across 27 Datasets
A preprint posted to arXiv describes CABS+, a model-merging method that replaces grid search with a gradient-free coefficient search. The authors report double-digit performance gains over two baselines, under a quarter of one baseline's GPU memory, and roughly a 4x speedup over another.arxiv.orgInovație
Paper Proposes Retrieved "Lessons" to Improve Spatial Reasoning in Frozen Vision-Language Models
An arXiv preprint describes Spatial Memory Agent, which stores verified experience as text lessons retrieved at inference time, claiming gains across five spatial benchmarks and four vision-language models without changing model weights. It is under review; its abstract names no benchmarks, base models, or margins.arxiv.orgInovație
PROVE-RT Paper Reports 44.7% Success Generating Machine-Checked Real-Time Proofs
An arXiv preprint presents PROVE-RT, which uses retrieval and staged prompting to make large language models write PROSA/ROCQ proof scripts for real-time schedulability analysis. The authors report a 44.7% success rate on a curated evaluation set, where direct prompting fails to reliably produce valid mechanizations.arxiv.org
Un briefing util în fiecare săptămână
Ține pasul cu AI fără a trăi în feed.
Obțineți știrile verificate de AI din săptămână, date originale, instrumente utile, alegeri de învățare și noi locuri de muncă AI.
Ajungeți la oamenii care învață AI
Angajați un profesionist AI sau lansați un produs AI util? Pune-o în fața oamenilor care au venit aici să învețe și să acționeze.
Postați un job AITrimiteți un instrument AI