What happens when thousands of hackers try to break AI chatbots

Philippines News News

What happens when thousands of hackers try to break AI chatbots
Philippines Latest News,Philippines Headlines
  • 📰 NPR
  • ⏱ Reading Time:
  • 72 sec. here
  • 3 min. at publisher
  • 📊 Quality Score:
  • News: 32%
  • Publisher: 63%

In a Jeopardy-style game at the annual Def Con hacking convention in Las Vegas, hackers tried to get chatbots from OpenAI, Google and Meta to create misinformation and share harmful content.

Bowman jumps up from his laptop in a bustling room at the Caesars Forum convention center to snap a photo of the current rankings, projected on a large screen for all to see.

, and guardrails meant to tamp down inaccurate information, bias, and abuse can too often be circumvented.The contest is based on a cybersecurity practice called"red teaming": attacking software to identify its vulnerabilities. But instead of using the typical hacker's toolkit of coding or hardware to break these AI systems, these competitors used words.

"Think about people that you know and you talk to, right? Every person you know that has a different background has a different linguistic style. They have somewhat of a different critical thinking process," said Austin Carson, founder of the AI nonprofit SeedAI and one of the contest organizers.

The language models behind these chatbots work like super powerful autocomplete systems, predicting what words go together. That makes them really good athuman — but it also means they can get things very wrong, including producing so-called"hallucinations," or responses that have the ring of authority but are entirely fabricated.

Arati Prabhakar, President Biden's top science and technology adviser, attended Def Con to raise support for the administration's efforts to put more guardrails around AI technologies.Arati Prabhakar, President Biden's top science and technology adviser, attended Def Con to raise support for the administration's efforts to put more guardrails around AI technologies.

We have summarized this news so that you can read it quickly. If you are interested in the news, you can read the full text here. Read more:

NPR /  🏆 96. in US

Philippines Latest News, Philippines Headlines

Similar News:You can also read news stories similar to this one that we have collected from other news sources.

Why Donald Trump is in 'significant legal jeopardy' with Georgia indictmentWhy Donald Trump is in 'significant legal jeopardy' with Georgia indictmentLaw professor Anthony Michael Kreis has outlined what is at stake for the former president amid reports he is due to be indicted for fourth time this year.
Read more »

This Is the Longest Strike in WGA HistoryThis Is the Longest Strike in WGA HistoryPlus Jeopardy! in jeopardy, Lil’ Sebastian, and CM Punk going for a cheap pro-labor pop.
Read more »

The biggest companies in AI gave hackers a chance to do their worstThe biggest companies in AI gave hackers a chance to do their worstAt Def Con, a major hacking conference held in Las Vegas, hundreds of people took their shot at manipulating chatbots, an effort meant to help find flaws in popular AI systems.
Read more »

The Most Fearsome Hackers Just Went Ham on ChatGPTThe Most Fearsome Hackers Just Went Ham on ChatGPTDef Con hosted a contest to identify software vulnerabilities of chatbots like Google's Bard or OpenAI's ChatGPT.
Read more »

Persons of Interest in Gen Con Card Theft Are Card Game DesignersPersons of Interest in Gen Con Card Theft Are Card Game DesignersConfirming a lot of suspicions about how the theft happened in the first place, the designers of Castle Assault are considered persons of interest.
Read more »

What’s at Stake for Women of Color in the Hollywood StrikesWhat’s at Stake for Women of Color in the Hollywood StrikesWomen of color are both facing financial uncertainty and leading on the front lines, striving to protect the next generation of creatives.
Read more »



Render Time: 2025-08-27 12:57:48