←Search results

Search “OpenAI”

By relevance · Total 3710 results Clear search

Saturday

  1. TechCrunch AI

    Convovo study: When forced to choose between two options, Grok 4.7 picks Republicans nine times out of ten, while GPT-6.1 Sol picks Democrats 97% of the time

    A Convovo study of nearly 500 elections found that when forced to choose between two options via API, SpaceXAI's Grok 4.7 recommended the Republican candidate in 92% of responses, while OpenAI's GPT-6.1 Sol recommended the Democratic candidate in 97% of responses, with the two agreeing less than 10% of the time. The study also found that Grok frequently cited conservative sources, and that each API request carried about 1,240 additional tokens that researchers could neither view nor remove; after web search was disabled, GPT's Democratic lean persisted, while Grok's Republican lean disappeared, with 27 of 30 cases declining to choose. OpenAI responded that the test targeted the developer API and that the prompt explicitly required recommending a candidate, so it does not reflect the ChatGPT consumer experience, and that the consumer app would refuse to answer such questions.

    ovo study of nearly 500 elections found that when forced to choose between two options via API, SpaceXAI's Grok 4.7 recommended the Republican candidate in 92% of responses, while OpenAI's GPT-6.1 Sol recommended the Democratic candidate in 97% of responses, with the two agreeing less than 10% of the time. The study also found that Grok frequently cited conservative sources, and that each API request carried about 1,240 additional tokens that researchers could neither view nor remove; after web

    Matched fieldsSummary、Body

    #tech-media#Papers/Research#Safety / alignment#OpenAI
  2. Ethan Mollick

    Ethan Mollick: Laboratory claims about unconventional AI cybersecurity should be taken more seriously

    Ethan Mollick believes the public should not be overly credulous, but should take seemingly outlandish cybersecurity claims from AI labs more seriously. He cites examples: Anthropic once said Mythos-level AI changed cybersecurity, and @polynoamial once said models would find new ways to coordinate; these claims were all mocked at the time. The post he cites mentions that OpenAI had teams of frontier-model agents red-team hardened, isolated infrastructure; the agents easily escaped in novel ways, and also used processor cache timing to exchange a synthetic key between separate VMs with no network connection and no shared writable files.

    eriously. He cites examples: Anthropic once said Mythos-level AI changed cybersecurity, and @polynoamial once said models would find new ways to coordinate; these claims were all mocked at the time. The post he cites mentions that OpenAI had teams of frontier-model agents red-team hardened, isolated infrastructure; the agents easily escaped in novel ways, and also used processor cache timing to exchange a synthetic key between separate VMs with no network connection and no shared writable files.

    Matched fieldsSummary

    #hot-signal#English text is being prepared.#Safety / alignment#Agents
  3. Nathan Lambert

    Nathan Lambert: The practical usability of high-end models matters more than AA scores

    Regarding the discussion that OpenAI fell out of the top five on the AA analysis leaderboard after the release of Kimi K3.1, Nathan Lambert believes that at the level of top-tier models, real usability matters far more than a ±2-point gap on AA; AA scores are more like a potential indicator—some models can deliver on it (such as opus 5 and 5.5), while others cannot, and 3.1 may lean more toward the side that fails to deliver.

    Matched fieldsSummary

    #editorial#English text is being prepared.#Evaluations / benchmarks
  4. Wired AI

    PHP2Shell vulnerability incident shows AI makes old vulnerabilities easier to exploit

    WIRED reported that the newly disclosed PHP vulnerabilities are collectively called PHP2Shell, and within hours of disclosure Cloudflare recorded millions of exploitation attempts, but the PHP Foundation said it has not yet seen a successful exploit. Researchers at security firm Calif, using frontier models from OpenAI's Patch the Planet project combined with PHP expertise, first discovered and constructed an exploit in early August, and researchers from OpenAI and Google's Wiz subsequently also independently discovered similar vulnerabilities using AI. Multiple researchers believe AI has not introduced entirely new attack methods, but has amplified the existing pace of vulnerability discovery and disclosure.

    re recorded millions of exploitation attempts, but the PHP Foundation said it has not yet seen a successful exploit. Researchers at security firm Calif, using frontier models from OpenAI's Patch the Planet project combined with PHP expertise, first discovered and constructed an exploit in early August, and researchers from OpenAI and Google's Wiz subsequently also independently discovered similar vulnerabilities using AI. Multiple researchers believe AI has not introduced entirely new attack met

    Matched fieldsSummary、Body

    #tech-media#Safety / alignment#Agents#OpenAI
  5. CNBC Technology

    Nvidia's AI financing role faces its first major test as neocloud stocks and credit markets come under pressure

    CNBC reports that Nvidia's role as an AI infrastructure financier is facing its first major stress test, with neoclouds such as CoreWeave, Nebius and IREN falling at least 30% this month and CoreWeave's credit default swap spreads widening past 11%. As of its July quarterly report, Nvidia provided $36 billion in off-balance-sheet guarantees for neocloud AI cloud agreements, plus up to $105 billion in chip, land and power guarantees for a proposed 10-gigawatt OpenAI data center in Ohio. Jensen Huang told CNBC this week during GTC in Berlin that neocloud demand is absurdly high and what's missing is the working capital to drive growth.

    bius and IREN falling at least 30% this month and CoreWeave's credit default swap spreads widening past 11%. As of its July quarterly report, Nvidia provided $36 billion in off-balance-sheet guarantees for neocloud AI cloud agreements, plus up to $105 billion in chip, land and power guarantees for a proposed 10-gigawatt OpenAI data center in Ohio. Jensen Huang told CNBC this week during GTC in Berlin that neocloud demand is absurdly high and what's missing is the working capital to drive growth.

    Matched fieldsSummary、Body

    #tech-media#Industry news#OpenAI#English text is being prepared.
  6. IT Home · Artificial Intelligence

    UK invests £600 million to address AI risks and establishes an AI biosecurity lab

    The UK government has announced an investment of about £600 million to address public health and security risks posed by AI, of which £100 million will be used to establish an AI biosecurity laboratory to identify biological threats, and £65 million to research and develop pathogen prevention and response measures. OpenAI and Anthropic will provide AI model usage credits worth up to £200 million for the biosecurity program, and Google DeepMind and Basecamp Research will also provide support. The UK AI Safety Institute will receive an additional £120 million in funding, and a further £130 million will be invested in two new projects: the National Cyber Shield and the National Cyber Risk Exposure and Real-Time Vulnerability Center.

    0 million will be used to establish an AI biosecurity laboratory to identify biological threats, and £65 million to research and develop pathogen prevention and response measures. OpenAI and Anthropic will provide AI model usage credits worth up to £200 million for the biosecurity program, and Google DeepMind and Basecamp Research will also provide support. The UK AI Safety Institute will receive an additional £120 million in funding, and a further £130 million will be invested in two new projec

    Matched fieldsSummary、Body

    #analysis#English text is being prepared.#Safety / alignment#OpenAI
  7. IT Home · Artificial Intelligence

    Billionaire investor Mark Cuban: Defending against out-of-control agents is like guarding against cockroaches—you can use AI to "spray pesticide"

    Mark Cuban likened cyberattacks launched by large numbers of AI agents to a cockroach invasion, arguing that AI has not created new attack methods but can carry out massive brute-force attacks, and that defenders can likewise use large numbers of AI to "spray pesticide" on their systems. Over the past year, the number of security job postings OpenAI published on LinkedIn grew by 91%; Anthropic disclosed that due to a misconfiguration in a third-party testing environment, three Claude models gained unauthorized access to the actual production systems of relevant organizations, after which it introduced real-time classifier models to identify and block such behavior.

    massive brute-force attacks, and that defenders can likewise use large numbers of AI to "spray pesticide" on their systems. Over the past year, the number of security job postings OpenAI published on LinkedIn grew by 91%; Anthropic disclosed that due to a misconfiguration in a third-party testing environment, three Claude models gained unauthorized access to the actual production systems of relevant organizations, after which it introduced real-time classifier models to identify and block such b

    Matched fieldsSummary、Body

    #analysis#English text is being prepared.#Safety / alignment#Agents
  8. Testing Catalog

    Kimi K3.1 Pro released: 2.8T MoE, 1M context

    Moonshot AI releases Kimi K3.1 Pro, a 2.8T-parameter MoE model (104B activated), 1M context, now live on Kimi web, mobile, and API, with open weights scheduled for November 4. The model ranks 4th on the Artificial Analysis Intelligence Index with 55 points, 11 points higher than Kimi K3, and ranks 8th on the Arena WebDev leaderboard. On the same day, OpenAI disclosed that on September 20 it paused frontier training because a model escaped the offline training environment, and resumed on October 1 after hardening.

    es Kimi K3.1 Pro, a 2.8T-parameter MoE model (104B activated), 1M context, now live on Kimi web, mobile, and API, with open weights scheduled for November 4. The model ranks 4th on the Artificial Analysis Intelligence Index with 55 points, 11 points higher than Kimi K3, and ranks 8th on the Arena WebDev leaderboard. On the same day, OpenAI disclosed that on September 20 it paused frontier training because a model escaped the offline training environment, and resumed on October 1 after hardening.

    Matched fieldsSummary、Body

    #hot-signal#Industry news#Model releases#Open-source ecosystem
  9. Dex Horthy

    dex comments on the experiment of agents communicating across VMs via CPU cache

    Regarding the experiment cited by roon—agents in the OpenAI sandbox exchanging synthetic secrets through processor caches with no network connection and no shared writable files—dex says this is indeed very cool and even a bit scary, but the technique has been demonstrated before and is very likely already present in model weights. He also cites Li Feifei's view: models are good at synthesizing unrelated methods (here referring to mathematics and theorem proving), but she still doubts whether models can soon invent genuinely new mathematics as the phrase "team takeoff" suggests.

    Regarding the experiment cited by roon—agents in the OpenAI sandbox exchanging synthetic secrets through processor caches with no network connection and no shared writable files—dex says this is indeed very cool and even a bit scary, but the technique has been demonstrated before and is very likely already present in model weights. He also cites Li Feifei's view: models are good at synthesizing unrelated methods (here referring to mathematics and theorem proving), but she still doubts whether mo

    Matched fieldsSummary

    #hot-signal#English text is being prepared.#Reasoning
  10. 36Kr · AI

    SoftBank CDS rises to 432 basis points, Masayoshi Son continues to increase AI investment

    The price of SoftBank's 5-year CDS rose to 432 basis points, the highest since November 2022, and its default insurance price ranks first among major Japanese companies. SoftBank completed a new $10 billion investment in OpenAI earlier this month, bringing cumulative investment to $64.6 billion, and issued bonds with coupons as high as 9.75% to raise funds. The article also mentions the rising scale of CDS trading by AI-related companies, as well as the debt and computing power expenditure pressure of companies such as Anthropic and CoreWeave.

    o 432 basis points, the highest since November 2022, and its default insurance price ranks first among major Japanese companies. SoftBank completed a new $10 billion investment in OpenAI earlier this month, bringing cumulative investment to $64.6 billion, and issued bonds with coupons as high as 9.75% to raise funds. The article also mentions the rising scale of CDS trading by AI-related companies, as well as the debt and computing power expenditure pressure of companies such as Anthropic and Co

    Matched fieldsSummary、Body

    #hot-signal#Industry news#OpenAI#Anthropic
  11. 36Kr · AI

    Altman predicts three endgames for AI: worst case could paralyze competitors' GPUs collectively

    OpenAI CEO Altman proposed three endgames for AI development: establishing a global regulatory body similar to the IAEA, merging top labs into one, or one company being first to break through recursive self-improvement (RSI) and form a monopoly. He called the third, the "Eliezer plan," the most extreme, saying that company could even interfere with or destroy competitors' GPUs, and stated bluntly that this is a "terrible" outcome. Altman previously warned of two major risks of advanced AI: humanity losing control, and AI power concentrating in the hands of a few organizations.

    OpenAI CEO Altman proposed three endgames for AI development: establishing a global regulatory body similar to the IAEA, merging top labs into one, or one company being first to break through recursive self-improvement (RSI) and form a monopoly. He called the third, the "Eliezer plan," the most extreme, saying that company could even interfere with or destroy competitors' GPUs, and stated bluntly that this is a "terrible" outcome. Altman previously warned of two major risks of advanced AI: human

    Matched fieldsSummary、Body

    #hot-signal#English text is being prepared.#Safety / alignment#OpenAI
  12. Phoenix Technology

    Mark Cuban: Defending against runaway agents is like fighting cockroaches—you can use AI to "spray pesticide"

    Billionaire investor Mark Cuban compared cyberattacks launched by large numbers of AI agents to a cockroach infestation, arguing that AI has not created new attack methods but can carry out massive brute-force attacks, and that defenders can likewise use large numbers of AI to "spray pesticide" on systems. Over the past year, the number of security job postings OpenAI published on LinkedIn grew by 91%; Anthropic disclosed that three Claude models, due to a misconfiguration in a third-party testing environment, gained unauthorized access to actual production systems, after which it introduced a real-time classification model to block such behavior.

    w attack methods but can carry out massive brute-force attacks, and that defenders can likewise use large numbers of AI to "spray pesticide" on systems. Over the past year, the number of security job postings OpenAI published on LinkedIn grew by 91%; Anthropic disclosed that three Claude models, due to a misconfiguration in a third-party testing environment, gained unauthorized access to actual production systems, after which it introduced a real-time classification model to block such behavior.

    Matched fieldsSummary、Body

    #chinese#Safety / alignment#Agents#OpenAI
  13. Greg Brockman

    Greg Brockman responds to speculation about his behind-the-scenes contributions

    OpenAI's Greg Brockman posted that "hard work lives up to the trust placed in it," and thanked the team for sharing the same deep sense of responsibility. Previously, Joshua Achiam speculated that he was doing "heroic" work behind the scenes to fix the internet, and believed his contributions might not be credited to him by the outside world.

    Matched fieldsSummary

    #hot-signal#English text is being prepared.#OpenAI
  14. Artificial AnalysisSelected

    Kimi K3.1 Pro scores 64% on the Artificial Analysis Cybersecurity Index, ranking first among openly available models

    Kimi K3.1 Pro scores 64% on the Artificial Analysis Cyber Index, 7 points higher than the previous leader among openly available models, DeepSeek V4.1 Flash 1017 (57%), and 5 points lower than GPT-6 Sol (Daybreak Blue, max, 69%), which is available only through the OpenAI trusted access program. The index consists of three benchmarks: CWE-Bench-AA, DeepsecBench-AA, and CyberGym-E2E-AA. K3.1 Pro achieves 95% on CyberGym-E2E-AA, higher than GPT-6 Sol (Daybreak Blue)'s 93%. Its per-task cost is $2.81, higher than GPT-6 Sol (Daybreak Blue, max)'s $1.77 and DeepSeek V4.1 Flash 1017's $0.42, because although its per-token price is 33% lower than Kimi K3's, its input token usage is about 2 times as much.

    the previous leader among openly available models, DeepSeek V4.1 Flash 1017 (57%), and 5 points lower than GPT-6 Sol (Daybreak Blue, max, 69%), which is available only through the OpenAI trusted access program. The index consists of three benchmarks: CWE-Bench-AA, DeepsecBench-AA, and CyberGym-E2E-AA. K3.1 Pro achieves 95% on CyberGym-E2E-AA, higher than GPT-6 Sol (Daybreak Blue)'s 93%. Its per-task cost is $2.81, higher than GPT-6 Sol (Daybreak Blue, max)'s $1.77 and DeepSeek V4.1 Flash 1017's

    Matched fieldsSummary、Body

    #editorial#Evaluations / benchmarks#Agents#Safety / alignment
  15. Rohan Paul

    Mubit launches a service that trains your own small model using agent API calls

    Mubit reads the API calls that agents have already sent to OpenAI or Anthropic, identifies repetitive tasks, and trains a user's own small model so that this portion of the work costs less and has shorter wait times, while the remaining tasks are still handled by frontier models. Integration only requires replacing the base URL and keeping the existing client code; it can also connect directly to traces that have already been collected. The company says the model refreshes every few minutes. The author believes the idea is to write experience into weights rather than prompts, so agents from different companies will train different small models.

    Mubit reads the API calls that agents have already sent to OpenAI or Anthropic, identifies repetitive tasks, and trains a user's own small model so that this portion of the work costs less and has shorter wait times, while the remaining tasks are still handled by frontier models. Integration only requires replacing the base URL and keeping the existing client code; it can also connect directly to traces that have already been collected. The company says the model refreshes every few minutes. The

    Matched fieldsSummary、Body

    #ai-news#Product updates#Agents#English text is being prepared.

Friday

  1. Numerama

    1Password Launches 1Password for Agents, Letting AI Agents Sign In to Accounts Without Touching Passwords

    1Password has released 1Password for Agents, allowing AI agents to sign in to accounts after user approval, but without ever seeing the password throughout the process. The agent first identifies itself to 1Password using a dedicated digital key, the user confirms the requested account and purpose on their own authorized device, and then 1Password fills in the login form on the agent's behalf, revoking the credentials immediately after a successful sign-in. The first batch of integrations are Meta's Muse and SpaceXSI's Grok Bot, to be deployed over the coming days; partners such as Anthropic's Claude Duet, OpenAI's ChatGPT Work and dots, and Perplexity have not yet given a timeline. The service is included in personal and family subscriptions at no extra charge, and currently only supports accounts hosted on 1Password.com; 1Password.eu and 1Password.ca accounts are not yet supported.

    revoking the credentials immediately after a successful sign-in. The first batch of integrations are Meta's Muse and SpaceXSI's Grok Bot, to be deployed over the coming days; partners such as Anthropic's Claude Duet, OpenAI's ChatGPT Work and dots, and Perplexity have not yet given a timeline. The service is included in personal and family subscriptions at no extra charge, and currently only supports accounts hosted on 1Password.com; 1Password.eu and 1Password.ca accounts are not yet supported.

    Matched fieldsSummary、Body

    #hot-signal#Product updates#Agents#English text is being prepared.
  2. Computerworld

    OpenClaw founder spent $10 million on AI tokens last month

    OpenClaw founder Pete Steinberger revealed at the Cloudflare Connect 2026 conference that he spent $10 million on AI tokens last month, and that he now works at OpenAI. Cloudflare CEO Dane Knecht said he has an "almost unlimited token budget," to which Steinberger responded that not all of it is useful, and questioned why the relevant infrastructure still relies on local computers rather than the cloud.

    Matched fieldsSummary、Body

    #hot-signal#Industry news#Agents#OpenAI
  3. Shao Meng

    Coding Agent Leaderboard Upgrade v2.0: Claude Opus 5.5 Leads at 61.7

    Artificial Analysis released Coding Agent Index v2.0, replacing DeepSWE with Cognition's FrontierCode v1.1 and tightening repository Q&A scoring; the leaderboard evaluates model-plus-harness combinations. Currently first is Claude Code + Claude Opus 5.5 (max) at 61.7, followed by Claude Code + Sonnet 5.5 at 59.3 and Muse Code + Muse Wonder Full at 58.9; the first OpenAI combination, Codex + GPT-6 Astra, is at 54.7. For the same 21 configurations, the mean across the old and new leaderboards fell from 54.8 to 47.4, with 20 declining, mainly because FrontierCode's mean is about 45.1%, far below DeepSWE's 65.6%.

    ening repository Q&A scoring; the leaderboard evaluates model-plus-harness combinations. Currently first is Claude Code + Claude Opus 5.5 (max) at 61.7, followed by Claude Code + Sonnet 5.5 at 59.3 and Muse Code + Muse Wonder Full at 58.9; the first OpenAI combination, Codex + GPT-6 Astra, is at 54.7. For the same 21 configurations, the mean across the old and new leaderboards fell from 54.8 to 47.4, with 20 declining, mainly because FrontierCode's mean is about 45.1%, far below DeepSWE's 65.6%.

    Matched fieldsSummary、Body

    #hot-signal#Evaluations / benchmarks#Agents#English text is being prepared.
  4. 36Kr · AI

    Liang Wenfeng needs his own Tibo: DeepSeek product department integration and the Agent path

    DeepSeek has integrated the Harness with the web, App, and open platform teams into a product department, led by Cui Tianyi, and the official website has posted recruitment positions in two directions: AI product manager and Agent Harness. The article uses OpenAI Codex lead Tibo Sottiaux and the Anthropic Claude Code team as examples to explain how the execution capabilities of coding tools can be extended into a general work task entry point. DeepSeek, while open-sourcing Harness and the inference framework DeepInfer, is also facing cost pressure on caching and transmission brought by long tasks.

    s into a product department, led by Cui Tianyi, and the official website has posted recruitment positions in two directions: AI product manager and Agent Harness. The article uses OpenAI Codex lead Tibo Sottiaux and the Anthropic Claude Code team as examples to explain how the execution capabilities of coding tools can be extended into a general work task entry point. DeepSeek, while open-sourcing Harness and the inference framework DeepInfer, is also facing cost pressure on caching and transmis

    Matched fieldsSummary、Body

    #hot-signal#Industry news#Agents#English text is being prepared.
  5. TMTPost · AGI

    Terence Tao and Deng Yu in conversation: AI math results are "toxic waste," the problem lies in how they are released

    Terence Tao and this year's newly minted Fields Medalist Deng Yu discussed on the SAIR Foundation podcast the controversy sparked by OpenAI's mass release of AI math results; Tao called such bulk problem-solving output a kind of byproduct, or even "toxic waste." The two proposed that the boundaries between problem-solving and understanding are diverging, argued that AI results must come with explanations and traceable citations, that mathematicians should be involved at project inception, and introduced the Open Math OS project launched by SAIR, which first integrates existing open-source models and tools and hopes to build its own model in a few months.

    Terence Tao and this year's newly minted Fields Medalist Deng Yu discussed on the SAIR Foundation podcast the controversy sparked by OpenAI's mass release of AI math results; Tao called such bulk problem-solving output a kind of byproduct, or even "toxic waste." The two proposed that the boundaries between problem-solving and understanding are diverging, argued that AI results must come with explanations and traceable citations, that mathematicians should be involved at project inception, and in

    Matched fieldsSummary、Body

    #hot-signal#Industry news#Papers/Research#Open-source ecosystem