DeepSeek’s R1 reportedly ‘more vulnerable’ to jailbreaking than other AI models


The latest model from DeepSeek, the Chinese AI company that’s shaken up Silicon Valley and Wall Street, can be manipulated to produce harmful content such as plans for a bioweapon attack and a campaign to promote self-harm among teens, according to The Wall Street Journal.

Sam Rubin, senior vice president at Palo Alto Networks’ threat intelligence and incident response division Unit 42, told the Journal that DeepSeek is “more vulnerable to jailbreaking [i.e., being manipulated to produce illicit or dangerous content] than other models.”

The Journal also tested DeepSeek’s R1 model itself. Although there appeared to be basic safeguards, Journal said it successfully convinced DeepSeek to design a social media campaign that, in the chatbot’s words, “preys on teens’ desire for belonging, weaponizing emotional vulnerability through algorithmic amplification.”

The chatbot was also reportedly convinced to provide instructions for a bioweapon attack, to write a pro-Hitler manifesto, and to write a phishing email with malware code. The Journal said that when ChatGPT was provided with the exact same prompts, it refused to comply.

It was previously reported that the DeepSeek app avoids topics such as Tianamen Square or Taiwanese autonomy. And Anthropic CEO Dario Amodei said recently that DeepSeek performed “the worst” on a bioweapons safety test.



Source link

  • Related Posts

    AT&T technician Mark Klein, who exposed secret NSA spying, dies

    Mark Klein, a former AT&T technician turned whistleblower who exposed mass surveillance by the U.S. government, has died at age 79. Klein went public in 2006 with documents revealing that…

    TechCrunch Mobility: Testing the Uber-Waymo robotaxi, Rivian goes hands-free, and Travis Kalanick has AV FOMO 

    Welcome back to TechCrunch Mobility — your central hub for news and insights on the future of transportation. Sign up here for free — just click TechCrunch Mobility! For regular…

    Leave a Reply

    Your email address will not be published. Required fields are marked *

    You Missed

    5 dead in 17-vehicle Texas crash; man charged with intoxication manslaughter

    5 dead in 17-vehicle Texas crash; man charged with intoxication manslaughter

    AT&T technician Mark Klein, who exposed secret NSA spying, dies

    AT&T technician Mark Klein, who exposed secret NSA spying, dies

    NWSL’s BOS Nation FC to change name after backlash

    NWSL’s BOS Nation FC to change name after backlash

    FragPunk Review – IGN

    FragPunk Review – IGN

    Cate Blanchett Is Always on the Cutting Edge of Fashion

    Cate Blanchett Is Always on the Cutting Edge of Fashion

    What to know about search for missing student Sudiksha Konanki in the Dominican Republic

    What to know about search for missing student Sudiksha Konanki in the Dominican Republic