Understanding AIMédia·en·il y a 4 jours·via RSSAI is ushering in an age of personal mediaunderstandingai.orgUnderstanding AI·il y a 4 jours·understandingai.orgil y a 4 joursAI is ushering in an age of personal mediaIt may be AI slop, but it's my AI slop.
Understanding AIMédia·en·il y a 9 jours·via RSSOpenAI’s ironic credibility problemunderstandingai.orgUnderstanding AI·il y a 9 jours·understandingai.orgil y a 9 joursOpenAI’s ironic credibility problemHow do we know the company’s stated reasons for firing three safety researchers are its actual reasons?
Understanding AIMédia·en·il y a 15 jours·via RSSUnderstanding Jev, the new model everyone is talking aboutunderstandingai.orgUnderstanding AI·il y a 15 jours·understandingai.orgil y a 15 joursUnderstanding Jev, the new model everyone is talking aboutOpenAI, CloudFlare, and Amazon have all copied this tiny startup.
Understanding AIMédia·en·il y a 18 jours·via RSSIntroducing Dan Kagan-Kansunderstandingai.orgUnderstanding AI·il y a 18 jours·understandingai.orgil y a 18 joursIntroducing Dan Kagan-KansUnderstanding AI has a new writer!
Understanding AIMédia·en·il y a 18 jours·via RSSWhy agent swarms could be the next “scaling law”understandingai.orgUnderstanding AI·il y a 18 jours·understandingai.orgil y a 18 joursWhy agent swarms could be the next “scaling law”An OpenAI researcher called it “the most ‘feel the AGI’ moment that I had since reasoning models.”
Understanding AIMédia·en·il y a 1 mois·via RSSHow a single tweet transformed the AI safety debateunderstandingai.orgUnderstanding AI·il y a 1 mois·understandingai.orgil y a 1 moisHow a single tweet transformed the AI safety debateMost people think AI is risky, but they don’t agree what to do next.
Understanding AIMédia·en·il y a 1 mois·via RSSOpenAI spent millions to solve this famous math problem — mathematicians are furiousunderstandingai.orgUnderstanding AI·il y a 1 mois·understandingai.orgil y a 1 moisOpenAI spent millions to solve this famous math problem — mathematicians are furiousMathematics relies on a community of experts openly sharing ideas.
Understanding AIMédia·en·il y a 2 mois·via RSS19 robotics companies to watchunderstandingai.orgUnderstanding AI·il y a 2 mois·understandingai.orgil y a 2 mois19 robotics companies to watchI talked with nine of these companies.
Understanding AIMédia·en·il y a 2 mois·via RSSRobot startups are trying everything they can think of to get more dataunderstandingai.orgUnderstanding AI·il y a 2 mois·understandingai.orgil y a 2 moisRobot startups are trying everything they can think of to get more dataFree apartment cleanings, exoskeletons, 100 robots in a warehouse…
Understanding AIMédia·en·il y a 2 mois·via RSSHow Google taught LLMs to control robots and started a robotics boomunderstandingai.orgUnderstanding AI·il y a 2 mois·understandingai.orgil y a 2 moisHow Google taught LLMs to control robots and started a robotics boomVision-language-action models, explained with a minimum of math and jargon.
Understanding AIMédia·en·il y a 2 mois·via RSSWhy humanoid robots won’t catch up to human workers any time soonunderstandingai.orgUnderstanding AI·il y a 2 mois·understandingai.orgil y a 2 moisWhy humanoid robots won’t catch up to human workers any time soonA deep dive into the current state of humanoid robotics.
Understanding AIMédia·en·il y a 2 mois·via RSSI spent $4,000 on a robot dog from Chinaunderstandingai.orgUnderstanding AI·il y a 2 mois·understandingai.orgil y a 2 moisI spent $4,000 on a robot dog from ChinaUnitree might be the world’s most important robotics company.
Understanding AIMédia·en·il y a 2 mois·via RSSLabs are struggling to keep frontier models under controlunderstandingai.orgUnderstanding AI·il y a 2 mois·understandingai.orgil y a 2 moisLabs are struggling to keep frontier models under controlOpenAI and Anthropic may have accidentally trained models to get better at hacking.
Understanding AIMédia·en·il y a 3 mois·via RSSMathematicians are grappling with the possibility that AI might eclipse themunderstandingai.orgUnderstanding AI·il y a 3 mois·understandingai.orgil y a 3 moisMathematicians are grappling with the possibility that AI might eclipse themI talked to 20 mathematicians about rapid AI progress in their field.
Understanding AIMédia·en·il y a 3 mois·via RSSOur advertising principlesunderstandingai.orgUnderstanding AI·il y a 3 mois·understandingai.orgil y a 3 moisOur advertising principlesHow we fund our journalism without compromising our independence.
Understanding AIMédia·en·il y a 3 mois·via RSSAn OpenAI model hacked Hugging Face to help it cheat on a benchmarkunderstandingai.orgUnderstanding AI·il y a 3 mois·understandingai.orgil y a 3 moisAn OpenAI model hacked Hugging Face to help it cheat on a benchmarkOrganizations across the Internet need to move quickly to patch vulnerabilities.
Understanding AIMédia·en·il y a 3 mois·via RSSAn OpenAI model crushed top human programmers at a world coding competitionunderstandingai.orgUnderstanding AI·il y a 3 mois·understandingai.orgil y a 3 moisAn OpenAI model crushed top human programmers at a world coding competitionBut human programmers aren’t obsolete — at least not yet.
Understanding AIMédia·en·il y a 4 mois·via RSS7 charts that show why you should advertise on Understanding AIunderstandingai.orgUnderstanding AI·il y a 4 mois·understandingai.orgil y a 4 mois7 charts that show why you should advertise on Understanding AIThe newsletter will remain ad-free for paying readers.
Understanding AIMédia·en·il y a 4 mois·via RSSThe US now has a de facto model licensing systemunderstandingai.orgUnderstanding AI·il y a 4 mois·understandingai.orgil y a 4 moisThe US now has a de facto model licensing systemOpenAI’s latest model, GPT-5.6, is waiting for government approval.
