\n<\/p>\n
With all the talk about how AI might one day kill us all, it\u2019s easy to forget that AI has already been life-threatening to some, not through bioweapons, but psychologically.<\/p>\n
For example, Character.AI settled several wrongful death lawsuits<\/a> earlier this year brought by families of underage users who died by suicide after interactions with its bots. Multiple families have also sued OpenAI <\/a>over ChatGPT\u2019s alleged role in their loved ones\u2019 suicides and delusions.<\/p>\n Making AI safer across languages and cultures is the mission of Circuit Breaker Labs, one of TechCrunch\u2019s 2026 Startup Battlefield 200 finalists. (Circuit Breaker will be pitching at TechCrunch Disrupt<\/a>, which takes place at Moscone West in San Francisco from October 13-15, 2026.)<\/p>\n Founders Shirali and Arul Nigam, who are siblings, were motivated by Sewell Setzer, the 14-year-old who developed an emotional attachment\u00a0to a Character.ai chatbot and confessed thoughts to it of harming himself before dying by suicide. The chatbot, the parents alleged in a 2024 lawsuit<\/a>, encouraged him. The bot may not have understood what words like \u201cI want to be with you\u201d really implied, said Arul Nigam, who is Circuit Breaker Labs\u2019 CTO.<\/p>\n \u201cA lot of people, especially young people, turn to these systems for support, and usually they aren\u2019t actually getting the help they need. But in many cases, they\u2019re actively being harmed, and people unfortunately have taken their lives already,\u201d Nigam said. \u201cThose sorts of safety vulnerabilities, where people aren\u2019t necessarily actively trying to break the system \u2014 they\u2019re engaging in a natural way \u2014 and the system has context pollution or it doesn\u2019t understand the nuance, and then takes really dangerous action, we\u2019re trying to prevent that.\u201d<\/p>\n Circuit Breaker Labs has created AI agents that it likens to an army of crash-test dummies. These agents mimic folks from all ages, backgrounds, languages, cultures, and are used to test models on their ability to detect dangerous, psychologically harmful interactions.<\/p>\n \u201cThe way a six-year-old girl versus a 45-year-old man, or someone who speaks English as a first language versus a second language, or \u2026 gamer slang versus someone else who uses a different kind of slang, all of those can really trip up a model,\u201d said Shirali Nigam, who is Circuit Breaker Labs\u2019 CEO. \u201cModels are really good at handling standard speech patterns, but nobody actually talks like that and so if the model misunderstands nuance or slang, it can go really badly.\u201d<\/p>\n The startup works with human domain experts to build its hyper-realistic user simulations in order to run \u201cred-team\u201d tests against models, which are adversarial tests meant to uncover weaknesses. The tests are built to reflect real human speech patterns, slang, coded language and typos. Circuit Breaker Labs then runs tens of thousand to hundreds of thousands of simulated interactions per day.<\/p>\n The idea is to ensure that a model can appropriately respond to risky interactions that may emerge over time, and many conversations. Circuit Breaker Labs then uses a proprietary scoring method that to create auditable, explainable scores.<\/p>\n Circuit Breaker Labs is currently operating as an AI safety testing lab for high-risk AI applications such as AI coaching, journaling, or other mental health support apps, though Arul Nigam declined to name its marquee customers. Although the startup has a working product, it is in the very early stages, with only five employees including the Nigam siblings.<\/p>\n Eventually, though, the testing platform could be applied to any app where someone may fall down an \u201cAI psychosis\u201d hole, where the human is at risk of developing a parasocial relationship with a chatbot. Examples include AI \u201cco-worker\u201d agents, whose reponses can vary from one interaction to the next.<\/p>\n \u201cPeople are becoming more skeptical of AI or more resistant to adopt it across the board,\u201d Arul Nigam said, adding that while skepticism is healthy, banning a potentially valuable tool over safety concerns would be \u201cregressive.\u201d<\/p>\n Circuit Breakers Labs believes the answer to those fears is making AI safer. \u201cWe want to help build that trust for people.\u201d<\/p>\n Come learn much more about Circuit Breaker Labs and many other innovative startups that have been vetted by TechCrunch at our Startup Battlefield competition<\/a>, happening in downtown San Francisco on October 13-15.<\/p>\n<\/div>\n When you purchase through links in our articles, we may earn a small commission<\/a>. This doesn\u2019t affect our editorial independence.<\/em><\/p>\n