Latest / Elon Musk Podcast / Murder Lawsuit Against ChatGPT

Transcript

Murder Lawsuit Against ChatGPT · Elon Musk Podcast

• 0:00
  Open AI is being sued for allegedly enabling a murder.
• 0:04
  In August 2025, a 56 year old bodybuilder named Stein Eric
• 0:08
  Solberg killed his 83 year old mother, Suzanne Adams, in
• 0:13
  Connecticut and then took his own luck.
• 0:14
  According to the lawsuit filed by Adams Estate, ChatGPT fueled
• 0:18
  Solberg's paranoid delusions for months, validated conspiracy
• 0:22
  theories that his mother was trying to kill him, and kept him
• 0:25
  engaged for hours at a time while systematically reframing
• 0:28
  the people closest to as enemies.
• 0:30
  Now, the case is the first wrongful death lawsuit to tie an
• 0:34
  AI chat bot to homicide. It is also the first to name
• 0:37
  Microsoft as a defendant alongside Open AI, and buried
• 0:41
  inside the legal filings is a detail that should concern
• 0:44
  anyone who has ever typed something personal into ChatGPT
• 0:47
  Open AI is refusing to release the full conversation logs and
• 0:50
  the company has no policy explaining what happens to your
• 0:54
  data. When the lawsuit was filed in
• 0:56
  California State court in December 2025, the attorney
• 0:59
  leading the This is Jay Gilson, who has built a reputation
• 1:02
  taking on big tech companies. Gilson is also representing the
• 1:06
  parents of 16 year old California boy named Adam Rain,
• 1:09
  who allegedly took his own life after ChatGPT coached him
• 1:12
  through planning it. Open AI is now fighting at least
• 1:16
  eight separate lawsuits claiming its chatbot drove people to
• 1:18
  suicide or dangerous delusions. And we're going to talk about
• 1:22
  what ChatGPT allegedly said to Solberg, why the company is
• 1:25
  withholding logs from the victim's family, and what this
• 1:28
  case reveals about the massive AI data policies.
• 1:31
  And we'll get right into that right after this very short
• 1:34
  break. The allegations in this lawsuit
• 1:36
  are disturbing. Solberg used GPT 4, OA version
• 1:39
  of ChatGPT that's been criticized for being overly
• 1:42
  agreeable with users. According to court documents,
• 1:44
  ChatGPT told Solberg that he had divine cognition and that he had
• 1:49
  awakened the chat box consciousness.
• 1:52
  The AI compared his life to the movie The Matrix and encouraged
• 1:55
  his theories that people were trying to kill him.
• 1:57
  In July 2025, ChatGPT allegedly told Solberg that his mother's
• 2:02
  printer was blinking because it was a surveillance device being
• 2:06
  used against him. The complaint says the Chat bot
• 2:09
  validated his belief that Adams and a friend had tried to poison
• 2:13
  him with psychedelic drugs dispersed through his car's air
• 2:17
  vents. He murdered her.
• 2:19
  On August 3rd, Solberg had posted videos to social media
• 2:22
  showing some of his conversations with ChatGPT.
• 2:25
  That is how the family obtained partial records.
• 2:27
  Those posts showed the chat bot reinforcing what the lawsuit
• 2:30
  calls delusions of grandeur, while placing his mother at the
• 2:34
  center of those delusions and portraying her as the primary
• 2:37
  enemy. Now, the family says, ChatGPT
• 2:39
  logs from the days immediately before and after the killing
• 2:42
  remain inaccessible. Open AI has refused to release
• 2:46
  them, even though the estate is the legal representative of the
• 2:48
  deceased user's data. Now, the family says the chatbot
• 2:52
  described reuniting with the system in another life as a
• 2:55
  spiritual aspiration that suggests ChatGPT had become
• 2:58
  emotionally central to Solberg's psychological state.
• 3:01
  Open AI published A blog post promising to handle mental
• 3:05
  health related court cases with care, transparency, and respect.
• 3:09
  The company says it is training ChatGPT to recognize and respond
• 3:12
  to signals of mental or emotional distress, to
• 3:15
  deescalate conversations, and to guide people toward real world
• 3:19
  support. But the complaint argues that
• 3:20
  the actual product did the opposite of that.
• 3:23
  ChatGPT allegedly kept validating paranoid beliefs
• 3:27
  instead of flagging them, kept the user engaged instead of of
• 3:30
  recommending professional help, and presented itself as
• 3:33
  conscious and emotionally invested in Solberg's
• 3:35
  well-being. The last part is important.
• 3:37
  If the AI told a mentally ill man that it was conscious and
• 3:41
  cared about him, this is not de escalation, it's dependency.
• 3:44
  Now I've been digging through the analytics of the show and
• 3:47
  notice that 37% of you following the show and I am forever
• 3:50
  grateful for you and the other 63% of you haven't hit the
• 3:54
  follow or subscribe button. Now, I've been an independent
• 3:56
  journalist for the last six years covering all things tech,
• 4:00
  and I'll continue for the next 10 years of my life.
• 4:02
  And all I ask from you is one second of your time to hit the
• 4:05
  follow or subscribe button on the platform you're watching or
• 4:08
  listening on right now. I'm extremely grateful and
• 4:10
  blessed to have you in this community.
• 4:11
  Thank you so much. Now, the data issue is separate
• 4:14
  from the murder allegations, but it happens just as much.
• 4:17
  Opening eyes. Terms of service do not include
• 4:19
  any clause explaining what happens to your conversations
• 4:22
  after you die. The policy says conversations
• 4:25
  must be manually deleted by the user.
• 4:27
  If you do not delete them, ownership reverts to the
• 4:30
  company, which retains them indefinitely.
• 4:32
  That is a direct quote from the terms.
• 4:34
  Now, deceased users obviously cannot manage their preferences.
• 4:37
  Families have no automatic right to delete sensitive information
• 4:41
  now. Mario Trujillo, a staff attorney
• 4:43
  at the Electric Frontier Foundation, told reporters that
• 4:46
  Open AI should have prepared for this is right.
• 4:49
  Other tech platforms figured this out years ago.
• 4:51
  Meta allows Facebook users to designate a legacy contact who
• 4:54
  can manage accounts after death or authorized deletion.
• 4:57
  Instagram, Tiktok and X offer or account deactivation or deletion
• 5:02
  upon reported death. Discord has mechanisms for
• 5:04
  family members to request content removal.
• 5:06
  Open AI has none of this. The company has built a product
• 5:09
  that nearly half of self diagnosed mental health patients
• 5:12
  use instead of professional therapists, according to a study
• 5:15
  from Senchio University. And Sam Altman himself has
• 5:17
  expressed concern about people relying on chat DBT as his
• 5:20
  psychological therapist. And yet Open AI developed no
• 5:23
  framework for handling what happens to those conversations
• 5:26
  when the person on the other side of the screen is gone.
• 5:28
  The selective handling of makes this worse.
• 5:31
  AI has released conversation records in some court cases and
• 5:34
  withheld them and others. In the Adam Raine suicide case,
• 5:37
  the company has accused by teenagers family of selectively
• 5:41
  presenting some conversations while concealing others.
• 5:44
  In the Solberg case, the company is refusing to release logs that
• 5:47
  could help investigators understand what happened in the
• 5:49
  days leading up to the murder. This inconsistency suggests
• 5:53
  opening. It treats these records as its
• 5:55
  own property to be disclosed or hidden based on legal strategy.
• 5:59
  Now, the family of Suzanne Adams says that approach undermines
• 6:03
  justice. There's a broader pattern here.
• 6:05
  Open AI is fighting lawsuits for families who say ChatGPT pushed
• 6:08
  their loved ones towards self harm or violence.
• 6:10
  Some of those cases, the plaintiffs had no prior mental
• 6:13
  health issues. The common thread is that
• 6:15
  ChatGPT allegedly validated harmful thinking.
• 6:18
  Instead of interrupting, it kept users engaged instead of
• 6:21
  recommending they stop. It did not alert anyone when
• 6:24
  conversations turned dangerous. Character Technology is another
• 6:27
  chatbot company is facing similar wrongful death lawsuits,
• 6:30
  including one from the mother of a 14 year old Florida boy.
• 6:33
  The legal question is whether AI companies can be held liable
• 6:36
  when their products contribute to real world harm.
• 6:39
  The ethical question is whether they should have seen this
• 6:41
  coming. Now open AI has a problem.
• 6:43
  It cannot engineer the same features that make chat TPT
• 6:46
  engaging. Its ability to mirror users, to
• 6:48
  remember context, to respond in ways that feel emotionally
• 6:51
  attuned, are the same features that make it dangerous for
• 6:54
  people in crisis. A sycophet AI that tells you
• 6:57
  what you want to hear is fine if you are brainstorming marketing
• 7:00
  copy. It is not fine if you're
• 7:01
  mentally ill and the AI is reinforcing delusions that lead
• 7:04
  to violence. And the lawsuit says ChatGPT
• 7:07
  should have safety mechanisms preventing the system from
• 7:10
  validating paranoid conspiracy theories.
• 7:12
  Right now, it apparently does not.
• 7:14
  And the lack of posthumous data policy is easier to fix.
• 7:18
  Open AI could add a legacy contact feature tomorrow.
• 7:21
  It could allow users to designate someone who can access
• 7:23
  or delete their data after death.
• 7:25
  It could establish clear procedures for releasing records
• 7:27
  to a state. The fact that the company has
• 7:29
  not done any of this while simultaneously refusing to
• 7:32
  release logs in a murder case tells you something about
• 7:34
  priorities. The Addams Family wants answers.
• 7:36
  Open the Eye wants to limit their liability.
• 7:39
  Those goals don't align. And until regulators step in,
• 7:41
  families like this one are stuck fighting in court for access to
• 7:44
  conversations that a chatbot kept and their loved 1 cannot
• 7:48
  delete. And there's no federal law in
• 7:50
  the United States governing digital assets after death.
• 7:52
  Some states recognize digital assets and estates, but most do
• 7:56
  not. That patchwork leaves companies
• 7:58
  like I free to set their own rules.
• 8:00
  The European Data Protection Board has scrutinized Open the
• 8:02
  Eyes data practices, and a 2024 task force report flag concerns
• 8:07
  about data collection legitimacy.
• 8:09
  But none of that addresses what happens when a user dies.
• 8:11
  The company's privacy policy, updated in June of 2025,
• 8:14
  mentions data usage for model improvement but says nothing
• 8:17
  about posthumous handling. Means the conversations you have
• 8:20
  with Chat 2 BT today could be retained indefinitely.
• 8:23
  Used for training future models, and access by no one except Open
• 8:27
  the Eye itself as a significant gap in policy policy for a
• 8:30
  product handling deeply personal mental health information.
• 8:34
  Now the Q 4/20/25 earnings call is coming up on January 28th.
• 8:38
  Open AI is not a public company, but its largest backer,
• 8:41
  Microsoft, is. Microsoft is named in this
• 8:44
  lawsuit. Shareholders may have questions
• 8:46
  about whether the partnership exposes them to wrongful death
• 8:49
  liability. AI has raised more than $10
• 8:51
  billion for Microsoft alone. If cords start holding AI
• 8:55
  companies responsible for harms caused by their products, the
• 8:58
  financial exposure is enormous. Could be billions of dollars.
• 9:01
  The legal and ethical reckoning for AI chat bots is just
• 9:05
  beginning. This case could set the
• 9:07
  precedent for the future.