Sam Altman, the CEO of Tucker Carlson's recent Openry, has drawn conversations about artificial intelligence into areas that rarely reach headlines. Instead of focusing on speculation about unemployment and machine consciousness, Carlson pushed Altman against subjects who could deliver immediate results in privacy, morality and safety.
Suicide and Mental Health
Carlson asked how AI should respond to users expressing suicidal thoughts. Altman has admitted that thousands of people in such states interact with ChatGpt every week. “There are 15,000 people who commit suicide per week,” he said. “About 10% of the world [is] I'm talking to chatgpt. This is like 1,500 people talking a week and still commit suicide at the end of it. ”
The questions became more keen when Carlson raised the question of supporting dying laws in a country like Canada. “If domestic laws are terminally ill, you can imagine a world where you need to present an option for this. Here is the laws of your country. He emphasizes the difference between a depressed teenager and a terminally ill patient, calling it a “big difference.”
Privacy and legal protection
Another focus was user privacy. Carlson asked if the government could request access to conversations with ChatGPT. Altman acknowledged what they could do and insisted on what he called “AI privilege.” “When you talk to AI about your medical history, legal issues, or legal advice, or what you're looking for these other things, I think the government owes a level of protection to the citizens out there, just as you would if you were talking to the human version of this,” he said.
Without such protection, private exchanges regarding health, financial or relationships may be open to state demands and commercial use. Altman said he has already pushed this in Washington and “I feel optimistic that the government will understand the importance of this and will be able to do it.”
The virtues of AI design
Carlson pushed Altman how moral rules are written in the system. Altman explained that responses are formed by “model specifications,” a document that defines what AI can say and cannot say. “We consulted like hundreds of moral philosophers, people who thought about the ethics of technology and systems,” he said. “Finally, we had to like to make some decisions.”
Carlson noted that these guidelines determine the framing of sensitive issues for billions of users. Altman accepted responsibility and said, “The person you think should be responsible for those calls is me. I'm the one who ultimately looks like a public face. I seem to be able to overturn those decisions and one of our boards.”
Military Use and Fatal Decisions
Carlson raised the prospects for military applications. He asked if Openai's technology could be used for operation. “I think a lot of people in the military are talking to ChatGpt for advice,” Altman replied. He denied plans to build autonomous weapons, but he admitted that he didn't know how he would feel about the role, saying, “I don't know exactly how I feel about it. I like our army. I'm very grateful that they keep us safe.”
Deepfakes and Identity
The interview also touched on deepfakes and the erosion of trust in digital media. Carlson warned that AI could make it impossible to separate from fake speeches or images from actual speeches without biometric authentication. Altman objected, saying, “I don't think biometrics are necessary or necessary to use technology. I don't think biometrics are essential.” He proposed an alternative to encrypted signatures or codewords.
Copyright and content disputes
Regarding the training data issue, Altman said Openai uses public information but would avoid avoiding reproducing copyrighted material in the output. “Models should not be plagiarized,” he said. “Models should not be able to plagiarize them the same way people can,” he said that users often complain that the system is too restrictive, and refused to display text that could still be protected.
The broader fear and unknown effects
Carlson urged Altman to acknowledge what keeps him awake at night. “Good night after ChatGpt is launched,” said Altman. After talking to the system, he noted not only users who die from suicide, but also unpredictable social impacts. “I've recently noticed that real people have picked up the unusual dictionary and rhythms of language models,” he said. “You have enough people to talk to the same model, and that actually causes change in behavior on a social scale.”
He also warned of the risks of biotechnology. “These models have a very good bio and can help design biological weapons,” he says, calling them one of his main concerns despite ongoing safety work.
What the exchange revealed
Conversations have shown that public discussions often center around work and AI sense issues, but more pressing issues receive little attention. Suicide prevention, legal protection for personal conversation, hidden moral norms, cultural drift, military applications, and the threat of deepfake all shape the future of AI use.
Carlson's question line forced Altman to directly address these overlooked risks. Exchanges revealed that while the most important challenge may not be attracting headlines, they determine how safe and substantially integrated AI is in everyday life.
Note: This post was edited/created using Genai Tools.
Read next: Samsung and Google's Still iPhone 17 Series on Battery Life and Durability
[ad_2]
Source link
