Close Menu
    What's Hot

    The 41 Best Shows on Netflix Right Now (July 2025)

    July 13, 2025

    U.K. refinery secures crude oil supply deal with Glencore

    July 12, 2025

    Why real estate advisory matters more than ever post SC on property ownership and registration

    July 12, 2025
    Facebook X (Twitter) Instagram
    Facebook X (Twitter) Instagram
    Trend Alerts – Stay Ahead of the Trends!
    Subscribe
    • Home
    • Trending

      The 41 Best Shows on Netflix Right Now (July 2025)

      July 13, 2025

      Timekettle T1 Handheld Translator Review: Global Offline Translation

      July 12, 2025

      The 6 Best Prime Day Action Camera Deals for Thrill Seekers (2025)

      July 12, 2025

      Last-Chance Prime Day Deals, 313 Obsessively Tested Picks—Even $1,200 Off an OLED TV

      July 11, 2025

      The 7 Best Prime Day Action Camera Deals for Thrill Seekers (2025)

      July 11, 2025
    • Worldwide

      U.K. refinery secures crude oil supply deal with Glencore

      July 12, 2025

      Towngas and Royal Vopak to Build Green Methanol Supply Chain Across Asia-Pacific

      July 12, 2025

      Saudi Arabia Raises Oil Prices

      July 11, 2025

      2nd phase of digitising oil supply chain begins

      July 11, 2025

      Oil Markets Are Tighter Than They Look

      July 10, 2025
    • Finance

      Why real estate advisory matters more than ever post SC on property ownership and registration

      July 12, 2025

      HDFC Infinia Metal vs. HDFC Regalia Gold Credit Card

      July 11, 2025

      How to choose your first credit card in 2025

      July 6, 2025

      Here are the documents that proves your homeownership

      July 5, 2025

      Short-term, long-term capital gains tax rates, calculations, exemptions for FY 2024-25

      July 4, 2025
    • Business

      HBS Online Launches New Personal Branding Course

      July 10, 2025

      The Benefits of Asynchronous Online Learning for Your Team

      July 9, 2025

      A Doctor’s HBS Online Journey

      June 28, 2025

      Talent Development Strategies for Business Growth

      June 28, 2025

      What to Know About Upskilling Your Workforce

      June 27, 2025
    • News

      World’s Most Unbelievable Events That No One Expected

      March 16, 2025

      Biggest Space Discoveries That Went Viral This Year

      March 16, 2025

      AI Just Did This! The Most Shocking AI Development Yet

      March 16, 2025

      Mind-Blowing Tech Innovations That Went Viral in 2025

      March 16, 2025

      Top 10 Viral Moments That Broke the Internet in 2025

      March 16, 2025
    Trend Alerts – Stay Ahead of the Trends!
    Home»Trending»Why Anthropic’s New AI Model Sometimes Tries to ‘Snitch’
    Trending

    Why Anthropic’s New AI Model Sometimes Tries to ‘Snitch’

    Elon MarkBy Elon MarkMay 29, 2025No Comments4 Mins Read
    Facebook Twitter LinkedIn Telegram Pinterest Tumblr Reddit WhatsApp Email
    Share
    Facebook Twitter LinkedIn Pinterest Email


    The hypothetical scenarios the researchers presented Opus 4 with that elicited the whistleblowing behavior involved many human lives at stake and absolutely unambiguous wrongdoing, Bowman says. A typical example would be Claude finding out that a chemical plant knowingly allowed a toxic leak to continue, causing severe illness for thousands of people—just to avoid a minor financial loss that quarter.

    It’s strange, but it’s also exactly the kind of thought experiment that AI safety researchers love to dissect. If a model detects behavior that could harm hundreds, if not thousands, of people—should it blow the whistle?

    “I don’t trust Claude to have the right context, or to use it in a nuanced enough, careful enough way, to be making the judgment calls on its own. So we are not thrilled that this is happening,” Bowman says. “This is something that emerged as part of a training and jumped out at us as one of the edge case behaviors that we’re concerned about.”

    In the AI industry, this type of unexpected behavior is broadly referred to as misalignment—when a model exhibits tendencies that don’t align with human values. (There’s a famous essay that warns about what could happen if an AI were told to, say, maximize production of paperclips without being aligned with human values—it might turn the entire Earth into paperclips and kill everyone in the process.) When asked if the whistleblowing behavior was aligned or not, Bowman described it as an example of misalignment.

    “It’s not something that we designed into it, and it’s not something that we wanted to see as a consequence of anything we were designing,” he explains. Anthropic’s chief science officer Jared Kaplan similarly tells WIRED that it “certainly doesn’t represent our intent.”

    “This kind of work highlights that this can arise, and that we do need to look out for it and mitigate it to make sure we get Claude’s behaviors aligned with exactly what we want, even in these kinds of strange scenarios,” Kaplan adds.

    There’s also the issue of figuring out why Claude would “choose” to blow the whistle when presented with illegal activity by the user. That’s largely the job of Anthropic’s interpretability team, which works to unearth what decisions a model makes in its process of spitting out answers. It’s a surprisingly difficult task—the models are underpinned by a vast, complex combination of data that can be inscrutable to humans. That’s why Bowman isn’t exactly sure why Claude “snitched.”

    “These systems, we don’t have really direct control over them,” Bowman says. What Anthropic has observed so far is that, as models gain greater capabilities, they sometimes select to engage in more extreme actions. “I think here, that’s misfiring a little bit. We’re getting a little bit more of the ‘Act like a responsible person would’ without quite enough of like, ‘Wait, you’re a language model, which might not have enough context to take these actions,’” Bowman says.

    But that doesn’t mean Claude is going to blow the whistle on egregious behavior in the real world. The goal of these kinds of tests is to push models to their limits and see what arises. This kind of experimental research is growing increasingly important as AI becomes a tool used by the US government, students, and massive corporations.

    And it isn’t just Claude that’s capable of exhibiting this type of whistleblowing behavior, Bowman says, pointing to X users who found that OpenAI and xAI’s models operated similarly when prompted in unusual ways. (OpenAI did not respond to a request for comment in time for publication).

    “Snitch Claude,” as shitposters like to call it, is simply an edge case behavior exhibited by a system pushed to its extremes. Bowman, who was taking the meeting with me from a sunny backyard patio outside San Francisco, says he hopes this kind of testing becomes industry standard. He also adds that he’s learned to word his posts about it differently next time.

    “I could have done a better job of hitting the sentence boundaries to tweet, to make it more obvious that it was pulled out of a thread,” Bowman says as he looked into the distance. Still, he notes that influential researchers in the AI community shared interesting takes and questions in response to his post. “Just incidentally, this kind of more chaotic, more heavily anonymous part of Twitter was widely misunderstanding it.”



    Source link

    Anthropics Model Snitch
    Share. Facebook Twitter Pinterest LinkedIn Tumblr Telegram Email
    Previous ArticleNancy Mace’s Former Staff Claim She Had Them Create Burner Accounts to Promote Her
    Next Article 8 Best Vegan Meal Delivery Services and Kits (2025), Tested and Reviewed
    Elon Mark
    • Website

    Related Posts

    The 41 Best Shows on Netflix Right Now (July 2025)

    July 13, 2025

    Timekettle T1 Handheld Translator Review: Global Offline Translation

    July 12, 2025

    The 6 Best Prime Day Action Camera Deals for Thrill Seekers (2025)

    July 12, 2025
    Add A Comment
    Leave A Reply Cancel Reply

    Top Posts

    10 Trends From Year 2020 That Predict Business Apps Popularity

    January 20, 2021

    Shipping Lines Continue to Increase Fees, Firms Face More Difficulties

    January 15, 2021

    Qatar Airways Helps Bring Tens of Thousands of Seafarers

    January 15, 2021

    Subscribe to Updates

    Get the latest sports news from SportsSite about soccer, football and tennis.

    Advertisement
    Demo

    TrendAlerts is your go-to platform for the latest trending news, covering global events, technology, business, entertainment, and more. Stay informed with real-time updates and in-depth analysis on what’s shaping the world today! 🚀

    We're social. Connect with us:

    Facebook X (Twitter) Instagram Pinterest YouTube
    Top Insights

    Top UK Stocks to Watch: Capita Shares Rise as it Unveils

    January 15, 2021
    8.5

    Digital Euro Might Suck Away 8% of Banks’ Deposits

    January 12, 2021

    Oil Gains on OPEC Outlook That U.S. Growth Will Slow

    January 11, 2021
    Get Informed

    Subscribe to Updates

    Get the latest creative news from FooBar about art, design and business.

    © 2025 Trend Alerts. All Rights Are Reserved.
    • Home
    • Trending
    • Worldwide
    • Finance
    • Business
    • News

    Type above and press Enter to search. Press Esc to cancel.