Agentic Misalignment: How LLMs could be insider threats

https://www.anthropic.com/research/agentic-misalignment

#HackerNews #AgenticMisalignment #LLMs #InsiderThreats #AIResearch #Anthropic #Cybersecurity

AbsenceBench: Language Models Can't Tell What's Missing

https://arxiv.org/abs/2506.11440

#HackerNews #AbsenceBench #LanguageModels #Missing #AIResearch #NaturalLanguageProcessing

Jürgen Schmidhuber：the Father of Generative AI Without Turing Award

http://www.jazzyear.com/article_info.html?id=1352

#HackerNews #JürgenSchmidhuber #GenerativeAI #TuringAward #AIResearch #MachineLearning

Extracting memorized pieces of books from open-weight language models

https://arxiv.org/abs/2505.12546

#HackerNews #Extracting #memorized #pieces #of #books #from #open-weight #language #models #languagemodels #AIresearch #bookextraction #openweightmodels #arxiv

How would you describe AI ?

Another "smart phone moment" ie we had no idea the damage it would cause to humanity till it was too late?

Or the chance to make the world a better place with scientific advances and freedom from monotony?

Please boost 👍🏻

#AiResearch #ai #auspol #technology #australianpolitics #australianews #regulateai #globalnews #auspol2025 #AusTech #ukpolitics #nzpol #eupolitics #canpol #uspol

OpenAI: l'IA può sviluppare una personalità propria
#AIResearch #AISafety #Etica #Futuro #IntelligenzaArtificiale #MachineLearning #OpenAI #Regolamentazione #TechNews

https://www.ceotech.it/openai-lia-puo-sviluppare-una-personalita-propria/

OpenAI Finds 'Toxicity Switch' Inside AI Models, Boosting Safety

#AI #OpenAI #AISafety #LLMs #AIEthics #AIResearch #MachineLearning #AIAlignment

https://winbuzzer.com/2025/06/19/openai-finds-toxicity-switch-inside-ai-models-boosting-safety-xcxwbn/

Altman vs. Zuckerberg: OpenAI CEO Accuses Meta of Poaching AI Experts With $100 Million Signing Bonuses

#AI #AITalentWar #OpenAI #Meta #SamAltman #Zuckerberg #BigTech #AIResearch #AIDevs

https://winbuzzer.com/2025/06/18/altman-vs-zuckerberg-openai-ceo-accuses-meta-of-poaching-ai-experts-with-100-million-signing-bonuses-xcxwbn/

AI data centres are already awash with sewage so why not

BBC News - 'Cool data centres with treated sewage', says Anglian Water
https://www.bbc.co.uk/news/articles/cwyx41kx8d5o

#AI #generativeAI #AiResearch

Real-time action chunking with large models

https://www.pi.website/research/real_time_chunking

#HackerNews #RealTimeActionChunking #LargeModels #AIResearch #MachineLearning #Innovation

Time Series Forecasting with Graph Transformers

https://kumo.ai/research/time-series-forecasting/

#HackerNews #TimeSeriesForecasting #GraphTransformers #AIResearch #MachineLearning #DataScience

A level-headed look at AI.

https://theconversation.com/will-ai-take-your-job-the-answer-could-hinge-on-the-4-ss-of-the-technologys-advantages-over-humans-258469

#AiResearch

> the LLM group's participants performed worse than their counterparts in the Brain-only group at all levels: neural, linguistic, scoring.

Your Brain on LLM: Accumulation of Cognitive Debt when Using an AI Assistant for Essay Writing Task
https://www.brainonllm.com/

#genAI #slop #creativeWriting #llm #AiResearch #science

A Framework for Characterizing Emergent Conflict Between Non-Coordinating Agents [pdf]

https://paperclipmaximizer.ai/Unaware_Adversaries.pdf

#HackerNews #EmergentConflict #NonCoordinatingAgents #AIResearch #GameTheory #PDF

"Brain-only participants exhibited the strongest, most distributed networks; Search Engine users showed moderate engagement; and LLM users displayed the weakest connectivity."

"LLM users also struggled to accurately quote their own work."

"Over four months, LLM users consistently underperformed at neural, linguistic, and behavioral levels."

https://arxiv.org/abs/2506.08872

@stefanschulz

#LLM #ai #aigen #noai #AiResearch #nolm #ki #kunstliche_intelligenz

Text-to-LoRA: Hypernetwork that generates task-specific LLM adapters (LoRAs)

https://github.com/SakanaAI/text-to-lora

#HackerNews #TextToLoRA #Hypernetwork #LLMAdapters #AIResearch #SakanaAI

¡Contraataque a Apple! Un nuevo paper rebate el polémico estudio “The Illusion of Thinking”, acusando a Apple de fallos de diseño en las pruebas. ¿Colapsan los modelos… o las evaluaciones? 🤔
#Apple #IA #LLM #AIResearch

Large Language Models Often Know When They Are Being Evaluated

https://arxiv.org/abs/2505.23836

#HackerNews #LargeLanguageModels #Evaluation #AIResearch #NLP #Insights

🚀🎮 "Revolutionary" newsflash: Q-learning is still gasping for air while the rest of AI freely dances around with scalable glory. Apparently, the memo didn't reach Park at UC Berkeley, who is still lost in some time-travel paradox from 2025. 🤖📉
https://seohong.me/blog/q-learning-is-not-yet-scalable/ #Qlearning #AIresearch #UCberkeley #TechNews #TimeTravel #HackerNews #ngated

Clinical knowledge in LLMs does not translate to human interactions

https://arxiv.org/pdf/2504.18919

#HackerNews #ClinicalKnowledge #LLMs #HumanInteractions #AIResearch #MachineLearning