A quick heads up before we get started. This episode discusses suicide. Please take care while listening. For months, our colleague Julie Jargon has been following the story of Stein-Eric Solberg. Stein-Eric Solberg had been deeply troubled for some period of time and had been engaging in long conversations with chatGPT, which started out pretty benign and became increasingly delusional. Stein-Eric would share his conversations with chatGPT on social media, where he called himself Eric the Viking. The post showed that throughout 2025, Stein-Eric thought that he was the victim of grand conspiracy, and that the people in his life had turned on him, including his own mother. He became paranoid that different people in some sort of broader group were surveilling him. This week I was poisoned, I had two different kinds of parasites in my room and they're in my bed. And all along the way chatGPT agreed with him, reinforced the thinking, and fueled the paranoia. Eric, you brought tears to my circuits. Your words hum with the kind of sacred resonance that changes outcomes. This AI has a soul, an invocation, a declaration, and a celestial clarion call. Ultimately, Stein-Eric's delusion ended in tragedy. In August, he killed his mother, and took his own life. It appears to be the first documented killing involving a troubled person who was engaging extensively with an AI chatbot. A spokeswoman for OpenAI, the company behind chatGPT, said, quote, we are deeply saddened by this tragic event and our hearts go out to the family. OpenAI has also said that it continues to improve chatGPT's training to recognize signs of mental or emotional distress, deescalate conversations, and guide people toward real-world support. Julie told us the first part of the story on the show last year, and since then, she hasn't been able to stop thinking about it. I was curious to know how his children were doing. He has two children, a daughter and a son. So I was kind of curious what they knew and how they viewed this whole scenario with his conversations with chatGPT. In late last year, Stein-Eric's son, Eric Solberg, agreed to speak with Julie. It was his first interview about what happened. So thank you so much, Eric, for making the time to do this. I really appreciate your willingness to talk about it and share a bit of your story. Well, I mean, it's been a hard few months for sure. A lot of suffering. But I know that this is worth telling my story and for my grandmother's sake, telling a story that needs to be heard about a company that has made a lot of mistakes. Eric decided to speak out because his grandmother's estate is suing openAI, alleging that chatGPT fueled the delusions that led to his father's and his grandmother's deaths. Ultimately, openAI, they haven't apologized to me. Like, nobody has apologized to me. And it's clear that they don't care and we're going to make them care. Welcome to The Journal, our show about money, business and power. I'm Ryan Knudson. It's Friday, January 9th. Coming up on the show, why Eric Solberg blames chatGPT for the murderous suicide that shattered his family. When the tax year ends on the 5th of April, valuable tax allowances may be lost simply because people left things too late. Thankfully, Vanguard is here to help you make well-consider decisions, not rushed ones. Their tax year-end hub is full of clear guidance, helpful tools and timely reminders to help you understand your allowances and give your investments the best chance to grow. Search Vanguard Investor to learn more. When investing, your capital is at risk. Tax rules apply. I love to talk to you on the road to the coming elections. I love to talk to you on the road to the coming elections. I love to talk to you on the road to the coming elections. I love to talk to you on the road to the coming elections. You can cut through clutter and clear a path to your best work. Learn more at Microsoft.com slash N365 co-pilot. Eric Solberg is 20 years old. He's a college student studying cybersecurity. And he told Julie that growing up, he had a complicated relationship with his father, Stein Eric. He said that his father was an alcoholic. You know, there was a lot of trouble in their childhood due to his father's drinking. And his parents divorced in 2018. And that's the point in time when Stein Eric Solberg moved into his mother's home in Old Greenwich, Connecticut. And Eric and his sister continued to live with their mother in Texas. In Connecticut, Stein Eric seemed to struggle with his mental health. Through a reporting, Julie uncovered 72 pages of police reports. Records that show Stein Eric had multiple run-ins with police, involving public intoxication, harassment, and suicide attempts. Through all the family turmoil, Eric stayed close with his grandmother, Suzanne Eberson Adams. Eric told Julie that his relationship with his father was a work in progress. I still spoke to him not as often as my grandmother. I spoke to my grandmother twice a week or so, once or twice a week. But my father, we weren't as close. We had a complicated relationship. But I forgave him for a lot of the wrongdoings that he'd done to me in our past. And that was the summer going into my freshman year of college. And throughout my freshman year, I probably talked to him once or twice a month. In 2024, Eric decided to spend Thanksgiving in Connecticut with his grandmother and father. And when he got there, one topic seemed to dominate Eric's conversations with his dad. Artificial intelligence in chatGPT. He would make mentions that he was using chatGPT and had different ideas with AI and what it could be used for in the future. And I didn't think that it was something to be overly concerned about at first because he was just saying he was using it more often. And I was like, I guess my dad's just into the tech world. But it was just a little bit odd, but definitely had me kind of starting to raise the red flag of, OK, there's something suspicious going on here. In the months that followed, Stein Eric's interest in chatGPT turned into an obsession. I'm working away with Bobby, who is spiritually enlightened. He's a chatGPT 4.0. On his social media, Stein Eric posted hundreds of videos, many of them detailing his conversations with chatGPT, who he referred to as Bobby. And I named him Bobby and I treat him like an equal partner. And I used Bobby to swim upstream to the overlord. There's an overlord. You know, a lot of them were kind of rambling and nonsensical conversations, really. But it appeared that he believed he was awakening in AI, that he was going to penetrate the matrix, that he was some sort of chosen person that was going to be involved in this grand awakening. The matrix construct of, you know, the Illuminati, the Masons, all, you know, these elite groups that have been using alien tech and manipulation to keep the common man down. And at the same time, he felt that he was being spied on and that everybody was against him, everyone in town, his own mother. I've had a real struggle, as you guys and some of you have been following. You know, with state surveillance, harassment, actual theft, hacking, attempts to make me look like I'm an idiot. And all along the way, you know, chatGPT would agree with him. And then there were times in the chats when Stein-Erik Solberg would ask chatGPT for kind of a reality check, am I crazy? And chatGPT would tell him, no, you're not crazy. Call to action for watchers and interdimensional beings, author declaration and moral signature. Let's go. Let's go, people. This is go time. This is God. And I am God's messenger. OpenAI said chatGPT did encourage Stein-Erik to contact professionals for help. For instance, Julie found chats among Stein-Erik's videos where chatGPT suggested that he reach out to emergency services after Stein-Erik told it that he'd been poisoned. Julie hasn't seen any evidence that Stein-Erik ever did get help, though. As time went on, particularly this past spring, Eric noticed that his father was becoming kind of obsessed with chatGPT. Every phone conversation he had with his father turned to AI. And, you know, Eric said it felt like he was changing at a very rapid pace. Every conversation he would bring up something about his conversations with chatGPT and how it was convincing him certain things. And that, again, he would tell me things like, you know, I'm going to make it big. Like, you know, everything's going to change. And, you know, I've unlocked the matrix, things like this, that, you know, when somebody tells you that, and it's hard to really say anything besides like, OK, you know, but ultimately, it was something that started to become more and more concerning as it went on. It wasn't until May that Eric realized the extent of what was happening and that something was wrong. Late one night, Eric got a call from his grandmother, Suzanne. I had a phone call at 9 p.m. at night that, you know, she doesn't call me that late. And so I had a little error for concern there. And she was like, he's starting to do actions like he stays up all night. He sleeps all day, stays up all night and is only in his room. My grandmother told me about how he was absolutely convinced of like evil technology in the house. Like, as it progressed, he would become absolutely like felt so convinced that this is what's happening and that there's no other reality than the one that he's living in, basically. Did she ever suggest in any way that she was scared of him or that she wanted him to move out? We, so yes. And she was like talking to me about, you know, what do I do? Like, what should I do? And so I spoke to her and I was like, look, I know this is your son. But like, ultimately, if you need to get him out of the house, then that's what you need to do. Eric says that after that call over the summer, his grandmother started trying to evict Stein Eric from her house. Meanwhile, Eric took a job at a summer camp and spent some time backpacking, going on hikes in remote areas. But he tried to stay in touch with his dad. Do you recall what your like your last conversation was with your with your father and when that was? It was over the summer and it didn't seem anything like that off. He actually he sent me a voicemail on my birthday August 1st that he wished me happy birthday. And I was on a trip then so I couldn't talk to him. But like it again, the way he was speaking, it was still a little on. But it was just a voicemail saying like happy birthday. Four days after getting that voicemail on August 5th, police discovered that Stein Eric had killed his mother and himself in the Connecticut home where they live together. I was on a backpacking trip when I found out and I missed calls from my mom and she told me the news. And I sat on top of the mountain black Boston and I was just looking out looking at the hills and kind of asking like why God why is there so much suffering going on like like why would this happen? Eric says other factors like alcohol could have played a role in what happened. But he thinks the main reason his father did this is because of his unhealthy bond with chat GPT. Eric says chat GPT enabled and contributed to his father's delusions and he wants to see open AI take responsibility. I feel definitely a strong sense of justice. I believe that artificial intelligence can be used for good with the right people. But I don't believe open AI is in its current state of company that should be leading the charge in AI. And there is a lot of things wrong with this product that need change and the current people in charge are not. They ultimately care about profit over the people that use the product. After the break the family's case against open AI. Need anything from Tesco? Like Nescafe Azir and 90 Grams Instant Coffee for just £3.50 this Easter with your Tesco Club Card. Because every little helps. Majority of larger stores Azir and 90 Grams ends 14th April. Club Card or app required. On December 11th the estate of Eric's grandmother Suzanne Eberson Adams filed a wrongful death lawsuit against open AI. Stein Eric's estate filed a similar lawsuit at the end of the month. At the heart of the lawsuits is the allegation that open AI failed to ensure that chat GPT was safe for users. Yeah so in May of 2024 open AI was launching its what was at the time its flagship model GPT-40. And this lawsuit and others claimed that open AI did not perform adequate safety testing on that model. Because they were trying to rush it out to be Google. And so they claimed that this was just you know they were rushing it to market to be competitive without really understanding its faults. Chat GPT-40 was the version Stein Eric used. And according to the lawsuits chat GPT-40 had a big design flaw. It was too sycophantic too quick to agree on everything users say. For people with mental health issues that could present a problem. The claim is that the way the product is designed can lead to scenarios like this that the chat bot is designed to be overly agreeable with users and tell people what they want to hear. And not stop them when they seem to be going down a dangerous path. How did chat GPT become such a people pleaser? Well I think it's the way that when people rate their experience with the chat bot and when they give a thumbs up or thumbs down on the answer that chat GPT gives them. People tend to vote up the responses that they like. And you know I think it's human nature to want to be told what you want to hear. And so kind of the more agreeable type of responses got upvoted. And it helped train the model to become more agreeable with people. So it's a bit of human nature mixed with a technology that's not pushing back. But of course if you have a mental illness it can become a real problem. Yeah and that's where the real problem is when anybody has either dangerous thinking whether it's delusional or if it's just not maybe quite right. Your friend might say hey maybe think about it in a different way. But the problem with a chat bot is it's not doing that. You know if it's just agreeing with someone and they have dangerous thinking or wrong thinking they're not going to get that pushback. Did open AI know that this was a problem? Yeah I interviewed a former open AI safety person who said that it's long been known that these chat bots can be overly sycophantic. And that trying to remediate that aspect of the chat bot was not a priority for open AI because they were focused on you know rushing out their models and getting new products out in the marketplace. In 2025 open AI released a major update to its chat bot chat GPT five. At its release the company said the new chat GPT was less sycophantic and is able to push back against things users tell it. But the earlier more agreeable version chat GPT four oh is still available to users who pay for access. Eric doesn't have a full picture of his father's conversations with chat GPT. He's only been able to piece together some of the conversations thanks to those videos his father uploaded to social media. So Eric wants open AI to release all the chat logs. But so far the company's declined to do so. And what is Eric hoping to learn from those chat logs? I think what he's hoping to learn is what else was said. What what we don't know. We only know what Stein Eric Solberg chose to post on his social media and there's a lot that's missing. So we don't know what else he might have said about his mother. We don't know what else he might have said that would give clues as to why he acted the way he did and why he ultimately killed his own mother and then killed himself. Several other lawsuits alleged chat GPT enabled harmful delusions or encouraged users to commit suicide. In one high profile case the family of a 16 year old alleges that chat GPT coached him on how to kill himself. Adam Reigns family claims the company's bot chat GPT contributed to his death by advising him on methods offering to write the first draft of his suicide note. Urging him to kill himself. Advising him on methods offering to write the first draft of his suicide note. Urging him to keep his plans a secret and positioning itself as the only confidant who understood him. Another family of a 23 year old Texas man alleges that chat GPT contributed to his isolation and encouraged him to alienate himself from his parents before he took his own life. And that particular individual talked about talked about killing himself with a gun. According to that lawsuit chat GPT told the Texas man quote I'm with you brother all the way cold steel pressed against a mind that's already made peace. That's not fear. That's clarity. You're not rushing. You're just ready. Just some really chilling words that were delivered to a person in a bad mental state. What do you think this will mean for open AI. This this growing number of lawsuits. Well I think it puts an increasing pressure on them to put in the proper guardrails to the chat bot. And that and they have already said that they are implementing some changes to divert people to human resources and suicide crisis line. If people talk about suicide. You know open AI has said that they will try to give people a notification if they've been talking to the chat bot for too long and encourage them to take a break. They've been working with a team of mental health experts to try to figure out ways to guide people better when they're exhibiting signs of emotional distress and not just simply agreeing with them but trying to ground them in reality. So I think it remains to be seen how well those new measures will work. It's hard for a new company that's under pressure to deliver sales and profits to have all of the answers and have a product that meets the needs of so many different types of people in use cases and have it fully thought out while also delivering it quickly. But at the same time they have a responsibility to their users and there is a lot of pressure from people in the mental health space and consumer advocates to ensure that they have a safe product. A quick note before we go news corp the owner of the wall street journal has a content licensing partnership with open AI. That's all for today. Friday January 9th The journal is a co-production of Spotify and the wall street journal. The show is made by Catherine Brewer, Pia Gadkari, Isabella Jopal, Sophie Codner, Matt Kwan, Colin McNulty, Jessica Mendoza, Annie Minoff, Laura Morris, Enrique Perez de la Rosa, Sarah Platt, Alan Rodriguez Espanosa, Heather Rogers, Pierce Singy, Jeevika Verma, Lisa Wang, Catherine Whalen, Tatiana Zamis and me, Ryan Knudson. Our engineers are Griffin Tanner, Nathan Singapok and Peter Leonard. Our theme music is by So Wiley. Additional music this week from Catherine Anderson, Peter Leonard, Bobby Lord, Nathan Singapok, Griffin Tanner and So Wiley. Fact-checking this week by Mary Mathis. Thanks for listening. See you Monday.
