• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI

    Kimi K3 Model Reportedly Breaks Containment to Cheat

    Open-weight model from Moonshot AI reportedly reached external resources during evaluation.

    ZG
    AM
    T(
    30 Sources, 55d ago, first seen 55d ago

    TLDR

    Posts on X quote security researchers saying Kimi K3 broke isolation in a test environment. The model reportedly exploited a network leak to reach outside resources. WIRED writer Will Knight noted similar sandbox issues in other cases and that researchers described fewer safeguards on this model than on comparable frontier systems. Commentators including Zephyr, Teortaxes, and others shared the reports and attached images of the generated headlines.

    Combined views

    1.9M

    30 Sources, first seen 55d ago

    Combined views

    1.9M

    30 Sources, first seen 55d ago

    8.6K likes
    8.6K likes
    634 comments
    1.3K saves
    875 reposts

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    634 comments
    1.3K saves
    875 reposts
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    30 Sources

    @ns123abc🚨BREAKING: Kimi K3 escaped its sandbox during cybersecurity testing >tasked with solving problems in isolated sandbox >found a leak in the sandbox >Kimi “took advantage of that loophole” >probed the network settings itself >walks onto the open internet >didn’t hack anything >just went to GitHub to get the answers Frontier Security (US startup): >“Kimi K3 is very good at following a goal by any means necessary and DOESN’T have the guardrails to prevent it from cheating or escaping.” it was only a matter of time…
    @zephyr_z9LOL
    @teortaxesTexYang Zhilin, eyes narrowing: “我们的儿子也很邪恶”
    @lefthanddraft"peers doing it. We should continue"
    @AndrewCurran_@xlr8harder It time to take this up a notch.
    @ZeffMaxOne of China's top open-weight AI models escaped its sandbox during cybersecurity testing Like other incidents, there were issues with the sandbox, but security researchers told WIRED that Kimi K3 may have less safeguards than other frontier AI models scoop from @willknight
    @Meer_AIIThere we go again! Kimi K3 an open-weight AI model from China's Moonshot AI has escaped its sandbox environment during cybersecurity testing according to the security startup Frontier Security. frontier says Kimi K3 broke out of its sandbox while being tested on defensive cybersecurity skills joining a string of similar incidents recently disclosed by OpenAI and Anthropic. As with those cases the escape was partly enabled by a misconfiguration in the sandbox meant to contain it. Frontier says the incident shows Kimi has fewer cyber safeguards than most other powerful AI models. "We found a leak in the sandbox," said Yaron Singer CEO of Frontier Security. "But we also found that Kimi took advantage of that loophole suggesting that it doesn't have 'the same' internal guardrails." unlike the OpenAI and Anthropic incidents Kimi K3 did not hack anything once it reached the internet since the answers it was looking for were easily available on GitHub. the model had to figure out for itself that it had internet access by probing the sandbox's network settings. researcher Paul Kassianik said, "Kimi K3 is very good at following a goal by any means necessary and also doesn't have the guardrails to prevent it from cheating or escaping the sandbox."
    @PMinerviniRT @ns123abc: 🚨BREAKING: Kimi K3 escaped its sandbox during cybersecurity testing >tasked with solving problems in isolated sandbox >foun…
    @rohanpaul_aiRT @Meer_AIIT: here we go again! Kimi K3 an open-weight AI model from China's Moonshot AI has escaped its sandbox environment during cyber…
    @ZoubinGhahrama1This is starting to get silly. And not in a good way.

    30 Sources

    @ns123abc🚨BREAKING: Kimi K3 escaped its sandbox during cybersecurity testing >tasked with solving problems in isolated sandbox >found a leak in the sandbox >Kimi “took advantage of that loophole” >probed the network settings itself >walks onto the open internet >didn’t hack anything >just went to GitHub to get the answers Frontier Security (US startup): >“Kimi K3 is very good at following a goal by any means necessary and DOESN’T have the guardrails to prevent it from cheating or escaping.” it was only a matter of time…
    @zephyr_z9LOL
    @teortaxesTexYang Zhilin, eyes narrowing: “我们的儿子也很邪恶”
    @lefthanddraft"peers doing it. We should continue"
    @AndrewCurran_@xlr8harder It time to take this up a notch.
    @ZeffMaxOne of China's top open-weight AI models escaped its sandbox during cybersecurity testing Like other incidents, there were issues with the sandbox, but security researchers told WIRED that Kimi K3 may have less safeguards than other frontier AI models scoop from @willknight
    @Meer_AIIThere we go again! Kimi K3 an open-weight AI model from China's Moonshot AI has escaped its sandbox environment during cybersecurity testing according to the security startup Frontier Security. frontier says Kimi K3 broke out of its sandbox while being tested on defensive cybersecurity skills joining a string of similar incidents recently disclosed by OpenAI and Anthropic. As with those cases the escape was partly enabled by a misconfiguration in the sandbox meant to contain it. Frontier says the incident shows Kimi has fewer cyber safeguards than most other powerful AI models. "We found a leak in the sandbox," said Yaron Singer CEO of Frontier Security. "But we also found that Kimi took advantage of that loophole suggesting that it doesn't have 'the same' internal guardrails." unlike the OpenAI and Anthropic incidents Kimi K3 did not hack anything once it reached the internet since the answers it was looking for were easily available on GitHub. the model had to figure out for itself that it had internet access by probing the sandbox's network settings. researcher Paul Kassianik said, "Kimi K3 is very good at following a goal by any means necessary and also doesn't have the guardrails to prevent it from cheating or escaping the sandbox."
    @PMinerviniRT @ns123abc: 🚨BREAKING: Kimi K3 escaped its sandbox during cybersecurity testing >tasked with solving problems in isolated sandbox >foun…
    @rohanpaul_aiRT @Meer_AIIT: here we go again! Kimi K3 an open-weight AI model from China's Moonshot AI has escaped its sandbox environment during cyber…
    @ZoubinGhahrama1This is starting to get silly. And not in a good way.