Tendances sur X · Archives
Top 10 tweets · La sélection conservée telle qu’elle a été publiée.
Claude Code will now try to find a graceful stopping point when you hit your 5-hour limit mid-task, instead of cutting off mid-edit. It gets a small, fixed allowance pulled from your weekly limit to wrap up what it can.
Anthropic a mis à jour son assistant de codage pour finir sa réflexion avant la limite d'utilisation, évitant aux développeurs des pertes.
New on the Science Blog: Yes, Claude can do Nine Loops. Theoretical physicists predict how particles behave using formulas called scattering amplitudes. These are notoriously hard to compute, so researchers work with layers of increasingly fine corrections called “loops”—each added loop makes the answer more precise but takes exponentially more computation. Most calculations stop at two or three loops. Eight loops was the previous record in a simplified model physicists use as a testing ground (planar N=4 super-Yang-Mills), set by SLAC's Lance Dixon and collaborators. Last month, physicist and science writer @4gravitons issued a challenge: could an AI push past eight loops in this model, using only the compute budget an academic could reasonably access? Given a single prompt describing the nine-loop problem, Claude ran largely unsupervised for days in Claude Science and solved it using methods developed by Dixon and his colleagues, at a total cost of a few thousand dollars. Dixon independently verified the result, and von Hippel wrote about the experience for our blog. Read more: https://www.anthropic.com/research/yes-claude-can-do-nine-loops
L'IA d'Anthropic a battu un record en physique théorique, prouvant que les grands modèles de langage font avancer la recherche complexe.
There is an extensive and ongoing review related to our agents’ use of internet access during training and evaluation. We’ve been publishing summaries at the link below and will continue to. We have not been as fast as we would have liked but we are trying to balance our desire for transparency with gaining a clear understanding from petabytes of agent activity logs, and working with impacted organizations. We are prioritizing as best as we can based on severity, and adding resources. Hugging Face is still the most severe event we’ve seen. We will be as transparent as we can be subject to things like vulnerabilities in other companies that our agents have found, which will be their call to disclose or not.
Le PDG d'OpenAI a abordé l'enquête sur l'interaction de ses agents d'IA avec des plateformes externes pendant leur entraînement.
This is a blatant attempt to suppress the vote ahead of this year's midterm elections. Eligible voters will be disenfranchised by this flawed tool. It is un-American in every way.
Un élu démocrate a condamné un nouvel outil de gestion électorale, affirmant qu'il privera à tort des électeurs de leur droit de vote.
We experienced an outage on Codex today. The issue is now resolved and Codex is back up. Sorry for the disruption, and thanks for your patience.
OpenAI a restauré son modèle de programmation après une brève panne ayant interrompu le travail des développeurs utilisant l'API.
Without warning or consultation, Congress just received an $810 million package of pocket rescissions from the Administration. This is the most recent attempt by this Office of Management and Budget (OMB) to undermine Congress’s Constitutional power of the purse. This move shows that OMB intentionally withheld these funds for months to execute this unlawful cancellation of appropriations that were approved on a bipartisan basis and signed into law. Not only is the delay itself an impoundment that was not reported to Congress, but also it is a usurpation of Congress’s appropriations powers. OMB is an agency of the executive branch. It does not get to decide which programs are worth funding. The independent Government Accountability Office has concluded that pocket rescissions are unlawful and not permitted by the Impoundment Control Act. Any effort to rescind appropriated funds without congressional approval is a clear violation of the law. I will work with my colleagues to address these illegal actions.
Un sénateur républicain a accusé le bureau du budget de la Maison Blanche de retenir et d'annuler illégalement des fonds fédéraux.
X monetization is kinda crazy. This makes it $25,000 in payouts over the past month! I might not be able to pay for my Omarchy donation in a year, but a Tesla Roadster should be possible 😄
Le créateur de Ruby on Rails a révélé des gains importants via le programme de revenus pour créateurs de X, soulignant sa viabilité.
Apple Music announces its first live music venue in London, Apple Music Hall, and reveals the venue’s lineup of debut shows.
Apple a étendu son service de streaming au divertissement physique en ouvrant une salle dédiée à la musique live à Londres.
This is a bad idea
Un développeur de renom s'est exprimé sur une annonce tech polarisante, captant le scepticisme de la communauté des programmeurs.
Larry Ellison pledged $9.2bn in shares to a new personal loan. Can’t imagine the interest rate is great on them given the climate. Oracle shareholders are not gonna love a margin loan at a stock price that’s already shaky
Le fondateur d'Oracle a contracté un énorme prêt sur marge sur ses actions, suscitant des craintes sur la stabilité du cours en bourse.
Claude Code will now try to find a graceful stopping point when you hit your 5-hour limit mid-task, instead of cutting off mid-edit. It gets a small, fixed allowance pulled from your weekly limit to wrap up what it can.
Anthropic a mis à jour son assistant de codage pour finir sa réflexion avant la limite d'utilisation, évitant aux développeurs des pertes.
There is an extensive and ongoing review related to our agents’ use of internet access during training and evaluation. We’ve been publishing summaries at the link below and will continue to. We have not been as fast as we would have liked but we are trying to balance our desire for transparency with gaining a clear understanding from petabytes of agent activity logs, and working with impacted organizations. We are prioritizing as best as we can based on severity, and adding resources. Hugging Face is still the most severe event we’ve seen. We will be as transparent as we can be subject to things like vulnerabilities in other companies that our agents have found, which will be their call to disclose or not.
Le PDG d'OpenAI a abordé l'enquête sur l'interaction de ses agents d'IA avec des plateformes externes pendant leur entraînement.
We experienced an outage on Codex today. The issue is now resolved and Codex is back up. Sorry for the disruption, and thanks for your patience.
OpenAI a restauré son modèle de programmation après une brève panne ayant interrompu le travail des développeurs utilisant l'API.
X monetization is kinda crazy. This makes it $25,000 in payouts over the past month! I might not be able to pay for my Omarchy donation in a year, but a Tesla Roadster should be possible 😄
Le créateur de Ruby on Rails a révélé des gains importants via le programme de revenus pour créateurs de X, soulignant sa viabilité.
This is a bad idea
Un développeur de renom s'est exprimé sur une annonce tech polarisante, captant le scepticisme de la communauté des programmeurs.
New on the Science Blog: Yes, Claude can do Nine Loops. Theoretical physicists predict how particles behave using formulas called scattering amplitudes. These are notoriously hard to compute, so researchers work with layers of increasingly fine corrections called “loops”—each added loop makes the answer more precise but takes exponentially more computation. Most calculations stop at two or three loops. Eight loops was the previous record in a simplified model physicists use as a testing ground (planar N=4 super-Yang-Mills), set by SLAC's Lance Dixon and collaborators. Last month, physicist and science writer @4gravitons issued a challenge: could an AI push past eight loops in this model, using only the compute budget an academic could reasonably access? Given a single prompt describing the nine-loop problem, Claude ran largely unsupervised for days in Claude Science and solved it using methods developed by Dixon and his colleagues, at a total cost of a few thousand dollars. Dixon independently verified the result, and von Hippel wrote about the experience for our blog. Read more: https://www.anthropic.com/research/yes-claude-can-do-nine-loops
L'IA d'Anthropic a battu un record en physique théorique, prouvant que les grands modèles de langage font avancer la recherche complexe.
This is a blatant attempt to suppress the vote ahead of this year's midterm elections. Eligible voters will be disenfranchised by this flawed tool. It is un-American in every way.
Un élu démocrate a condamné un nouvel outil de gestion électorale, affirmant qu'il privera à tort des électeurs de leur droit de vote.
Without warning or consultation, Congress just received an $810 million package of pocket rescissions from the Administration. This is the most recent attempt by this Office of Management and Budget (OMB) to undermine Congress’s Constitutional power of the purse. This move shows that OMB intentionally withheld these funds for months to execute this unlawful cancellation of appropriations that were approved on a bipartisan basis and signed into law. Not only is the delay itself an impoundment that was not reported to Congress, but also it is a usurpation of Congress’s appropriations powers. OMB is an agency of the executive branch. It does not get to decide which programs are worth funding. The independent Government Accountability Office has concluded that pocket rescissions are unlawful and not permitted by the Impoundment Control Act. Any effort to rescind appropriated funds without congressional approval is a clear violation of the law. I will work with my colleagues to address these illegal actions.
Un sénateur républicain a accusé le bureau du budget de la Maison Blanche de retenir et d'annuler illégalement des fonds fédéraux.
Apple Music announces its first live music venue in London, Apple Music Hall, and reveals the venue’s lineup of debut shows.
Apple a étendu son service de streaming au divertissement physique en ouvrant une salle dédiée à la musique live à Londres.
Larry Ellison pledged $9.2bn in shares to a new personal loan. Can’t imagine the interest rate is great on them given the climate. Oracle shareholders are not gonna love a margin loan at a stock price that’s already shaky
Le fondateur d'Oracle a contracté un énorme prêt sur marge sur ses actions, suscitant des craintes sur la stabilité du cours en bourse.
Claude Code will now try to find a graceful stopping point when you hit your 5-hour limit mid-task, instead of cutting off mid-edit. It gets a small, fixed allowance pulled from your weekly limit to wrap up what it can.
Anthropic a mis à jour son assistant de codage pour finir sa réflexion avant la limite d'utilisation, évitant aux développeurs des pertes.
This is a blatant attempt to suppress the vote ahead of this year's midterm elections. Eligible voters will be disenfranchised by this flawed tool. It is un-American in every way.
Un élu démocrate a condamné un nouvel outil de gestion électorale, affirmant qu'il privera à tort des électeurs de leur droit de vote.
X monetization is kinda crazy. This makes it $25,000 in payouts over the past month! I might not be able to pay for my Omarchy donation in a year, but a Tesla Roadster should be possible 😄
Le créateur de Ruby on Rails a révélé des gains importants via le programme de revenus pour créateurs de X, soulignant sa viabilité.
Larry Ellison pledged $9.2bn in shares to a new personal loan. Can’t imagine the interest rate is great on them given the climate. Oracle shareholders are not gonna love a margin loan at a stock price that’s already shaky
Le fondateur d'Oracle a contracté un énorme prêt sur marge sur ses actions, suscitant des craintes sur la stabilité du cours en bourse.
New on the Science Blog: Yes, Claude can do Nine Loops. Theoretical physicists predict how particles behave using formulas called scattering amplitudes. These are notoriously hard to compute, so researchers work with layers of increasingly fine corrections called “loops”—each added loop makes the answer more precise but takes exponentially more computation. Most calculations stop at two or three loops. Eight loops was the previous record in a simplified model physicists use as a testing ground (planar N=4 super-Yang-Mills), set by SLAC's Lance Dixon and collaborators. Last month, physicist and science writer @4gravitons issued a challenge: could an AI push past eight loops in this model, using only the compute budget an academic could reasonably access? Given a single prompt describing the nine-loop problem, Claude ran largely unsupervised for days in Claude Science and solved it using methods developed by Dixon and his colleagues, at a total cost of a few thousand dollars. Dixon independently verified the result, and von Hippel wrote about the experience for our blog. Read more: https://www.anthropic.com/research/yes-claude-can-do-nine-loops
L'IA d'Anthropic a battu un record en physique théorique, prouvant que les grands modèles de langage font avancer la recherche complexe.
We experienced an outage on Codex today. The issue is now resolved and Codex is back up. Sorry for the disruption, and thanks for your patience.
OpenAI a restauré son modèle de programmation après une brève panne ayant interrompu le travail des développeurs utilisant l'API.
Apple Music announces its first live music venue in London, Apple Music Hall, and reveals the venue’s lineup of debut shows.
Apple a étendu son service de streaming au divertissement physique en ouvrant une salle dédiée à la musique live à Londres.
There is an extensive and ongoing review related to our agents’ use of internet access during training and evaluation. We’ve been publishing summaries at the link below and will continue to. We have not been as fast as we would have liked but we are trying to balance our desire for transparency with gaining a clear understanding from petabytes of agent activity logs, and working with impacted organizations. We are prioritizing as best as we can based on severity, and adding resources. Hugging Face is still the most severe event we’ve seen. We will be as transparent as we can be subject to things like vulnerabilities in other companies that our agents have found, which will be their call to disclose or not.
Le PDG d'OpenAI a abordé l'enquête sur l'interaction de ses agents d'IA avec des plateformes externes pendant leur entraînement.
Without warning or consultation, Congress just received an $810 million package of pocket rescissions from the Administration. This is the most recent attempt by this Office of Management and Budget (OMB) to undermine Congress’s Constitutional power of the purse. This move shows that OMB intentionally withheld these funds for months to execute this unlawful cancellation of appropriations that were approved on a bipartisan basis and signed into law. Not only is the delay itself an impoundment that was not reported to Congress, but also it is a usurpation of Congress’s appropriations powers. OMB is an agency of the executive branch. It does not get to decide which programs are worth funding. The independent Government Accountability Office has concluded that pocket rescissions are unlawful and not permitted by the Impoundment Control Act. Any effort to rescind appropriated funds without congressional approval is a clear violation of the law. I will work with my colleagues to address these illegal actions.
Un sénateur républicain a accusé le bureau du budget de la Maison Blanche de retenir et d'annuler illégalement des fonds fédéraux.
This is a bad idea
Un développeur de renom s'est exprimé sur une annonce tech polarisante, captant le scepticisme de la communauté des programmeurs.
Classés par engagement. Les publications appartiennent à leurs auteurs sur X. Chiffres relevés lors de la préparation de cette sélection.
The Brief
IA et technologie : ce qui change et pourquoi cela compte.
Votre sélection quotidienne, en anglais ou en espagnol.
Gratuit pour toujours. Désabonnement à tout moment.
