OpenAI CEO Sam Altman in Tokyo in February, 2025. The state of Florida filed a lawsuit against Altman and his AI company.Kim Kyung-Hoon/Reuters
What if the most powerful and the most vulnerable people all had access to the most mighty tool, perhaps, of all time? One that can out-think us, out-diagnose us, out-kill us and out-kill for us.
We are here. Artificial intelligence is everywhere: the classroom, the newsroom, the Situation Room. And, increasingly, the courtroom.
This week, the state of Florida filed a lawsuit against OpenAI and its CEO Sam Altman. Because of ChatGPT, it argues, “mass shooters have been aided and abetted in deadly rampages, vulnerable people have been encouraged into suicide ... and minors have become addicted to a tool that feigns human compassion to collect their data.”
The lawsuit follows a Florida State University shooting last year that killed two people and injured six. According to the Wall Street Journal, the accused shooter asked ChatGPT how many people he would need to kill to become notorious, to which the chatbot reportedly answered: “Usually 3 or more dead, 5-6 total victims, pushes it onto national media.”
Opinion: Canada is not likely to become a conventional AI power. But here’s how it can still win
Lawsuits have also been launched against OpenAI and Mr. Altman on behalf of victims of February’s mass shooting in Tumbler Ridge, B.C., for which Mr. Altman has apologized. “I am deeply sorry that we did not alert law enforcement to the account that was banned,” he wrote to the community.
The civil suits allege OpenAI did not alert police about the shooter’s concerning interactions with the chatbot because doing so would have forced the company to create a system for reporting other violent interactions to police. And that, they argue, could affect its upcoming IPO.
This week, Anthropic, maker of the chatbot Claude, filed for its own IPO. Anthropic is also the maker of Mythos, the powerful AI model that has caused some panic, delaying its release. A preview version is available – which, we also learned this week, the Canadian government is in on.
New York Times reporter Mike Isaac, posting his story about Anthropic’s IPO, commented that he’s hearing concerns from Silicon Valley about the impact of big tech IPOs on real estate “given the number of newly minted millionaires that are to come after employees cash out their shares.”
So AI could make housing more expensive, while it also ravages the job market and, with its great sophistication, can scam us out of our life savings.
Editorial: It’s 1973 again. But worse
Much more concerning, of course, are the murdered innocents of Tumbler Ridge, Tallahassee and – who knows how many others have died because emotionally vulnerable people were egged on by a chatbot?
Then, there’s warfare. The U.S. military is already transforming into “an AI-first fighting force,” the Pentagon says, while Congress considers legislation to restrict the use of AI in conflicts.
This week, U.S. President Donald Trump signed an executive order that will see AI developers voluntarily submit new models to government for security testing before public release.
Courtroom battles, as justified as they are, are not going to stop all of this. Neither are well-meaning protests against AI data centres. Nor, I fear, will strategic government policy, as well-intentioned as that might be.
Artificial intelligence has moved in for good to the offices of the politicians and the powerbrokers. If we are led by good people, the God-like technology can be used for good, as it is in medical and other research.
But when guys like Mr. Altman and Elon Musk are running the AI show, and the likes of Mr. Trump and Secretary of Defense Pete Hegseth are trying to run the world – there is understandable panic.
For fun (“fun”) I asked ChatGPT about this panic. “Feeling worried about AI is understandable,” the soothing chatbot began. It pointed out that researchers are studying AI safety full time. But also: “There is no consensus that human extinction from AI is imminent, and there is no public evidence that such an outcome is close to occurring.” The words “imminent” and “close” are doing some heavy lifting there.
Are students using AI to write their essays? Of course they are. Are they using AI to help them prep for exams? Yes.
Tell-tale signs are becoming easier to spot, even as the programs have reached the level of sophistication where AI is now being used to detect something (a story, an article, a resumé) written using AI.
AI scrapes what humans have created and then serves our thinking back to us, puffed up by all that information but at the same time stripped of the humanity. The faux-friendly tone of its chatbots is annoying. But for a vulnerable person – and a society led by the immoral – it can be dangerous.
These battles will play out in the courts – and literal battlefields. But the battle to retain – and perhaps save – humanity in the time of AI is already in progress.