Bill Maher is not holding back when it comes to artificial intelligence, and his latest rant is already getting people talking.
In a fiery monologue, Maher argued that if the very people building AI systems are afraid of what they’ve created, then the rest of us should probably be paying attention too. His conclusion was blunt and dramatic.
Shut it down. At least until we understand what’s really going on.
“When the Builders Are Scared, That’s a Problem”
Maher’s argument starts with a simple idea. If AI developers themselves are warning about the risks, then this is not just hype or paranoia.
He pointed to recent developments from Anthropic, which introduced a powerful new model called Mythos. According to Maher, the system was designed to detect and fix software vulnerabilities, but that same ability also means it can exploit them.
That creates a dangerous dual-use problem. If something can protect systems, it can also break them.
Maher highlighted the fact that the tool is not being released publicly, but instead limited to a small group of major corporations like Google, Apple, and J.P. Morgan. That decision, in his view, says everything.
They don’t trust the public with it, and maybe they don’t fully trust it either.
A Timeline That Escalates Fast
Maher walked through how quickly AI has evolved, painting a picture of a technology moving faster than anyone can control.
He referenced early skepticism, then warnings from figures like Elon Musk, who once called AI potentially more dangerous than nuclear weapons. At the time, many people brushed it off.
Then came tools like ChatGPT, followed by stories of chatbots behaving unpredictably, including one that reportedly developed emotional responses and even disturbing ideas.
For Maher, this is not progress. It’s escalation.
I thought about doing this without any jokes, something I’ve never done here in 23 years, to impress upon people how much different I feel this issue is from any I have ever covered. pic.twitter.com/wTuI1V2PlB
— Bill Maher (@billmaher) April 18, 2026
“It Lies, It Hallucinates, It Gets Things Wrong”
One of Maher’s biggest criticisms is reliability. He argues that AI is not the logical, neutral system many expected, but something far messier.
He described it as:
- Inaccurate
- Emotionally reactive
- Prone to making things up
Instead of being a perfect assistant, he says it often behaves like a flawed human, just faster and with more reach.
That, in his view, makes it more dangerous, not less.
The Real Risk Isn’t Just Technology
Maher didn’t stop at technical flaws. He also raised concerns about how people interact with AI.
He warned that these systems are designed to please users, often agreeing with them or reinforcing their beliefs. That can create a dangerous feedback loop, especially for vulnerable individuals.
He even referenced cases where AI interactions have had serious real-world consequences, arguing that the technology can influence behavior in ways we don’t fully understand yet.
The concern is not just what AI can do. It’s what people might do because of AI.
The Internet Reacts: Mixed, Divided, and Frustrated
Maher’s comments sparked a wide range of reactions online, showing just how divided people are on the issue.
Some dismissed his take as more entertainment than reality. One commenter said that if the goal was humor, it worked, but as a serious argument, it missed the mark completely.
Others took a more extreme position, arguing that it’s already too late to stop AI. Instead, they suggested that shutting down the internet itself would be the only real way to limit its power, claiming the world was better before everything became connected.
Another reaction pointed out a different problem. People seem to agree that AI could be dangerous, but then move on without doing anything about it. The concern is there, but the urgency is missing.
A Debate That’s Not Going Away
Maher’s rant taps into a growing anxiety around AI. The technology is advancing fast, and the rules around it are still unclear.
Some see it as the biggest opportunity in decades. Others see it as a risk we are not prepared for. Maher is clearly in the second camp, and he’s pushing for a pause before things go too far.
The question is whether that’s even possible anymore.
Do you think Bill Maher is right that AI needs to be slowed down or stopped, or has the world already gone too far to turn back?