Reddit's co-founder, Alexis Ohanian, stated on Wednesday that the technology industry has been "tone-deaf" in its efforts to explain artificial intelligence and properly inform the public about its genuine risks. Ohanian argued that the debate surrounding AI should center on substantive dangers rather than topics that can be leveraged for quick political gains, emphasizing that an informed public is vital for navigating this technology successfully.

He conceded that the tech sector has exacerbated the problem, admitting it has done a "pretty tone-deaf" job of explaining AI, while noting its risks are more mundane than the "Terminator and Skynet" scenarios some fear. Skynet, of course, is the fictional self-aware AI system from the Terminator films that views humanity as a threat and attempts to eradicate it.

These comments arrive just days after researchers at AI labs OpenAI and Anthropic issued warnings about the rapid advancement of a technique known as recursive self-improvement (RSI). RSI refers to AI systems helping to develop more advanced models, potentially creating a loop where increasingly capable systems build even more capable ones. The concern is that if AI systems take over the training process for new models, their human developers could ultimately lose control over how these systems evolve.

While acknowledging that AI risks are real, Ohanian believes that doomsday-style "Terminator" warnings might distract from more tangible concerns, such as agent swarms—where multiple AI agents collaborate to execute complex tasks. "There are risks, and I do think they're a lot more benign, or more mundane, than people think," he said. He drew a comparison between AI and nuclear energy, noting both are powerful scientific advancements that can cause severe harm or deliver significant benefits depending on how responsibly they are managed.

"I want this to go in a healthy direction, and that just requires fewer tweets and more thoughtful conversations," Ohanian said. "Doom-posting is demoralizing," he added, without naming specifics, saying it is "frustrating" when "the slogans aren't actually the real thing."

Earlier this month, an AI researcher resigned from Anthropic, accusing both that company and OpenAI of behaving irresponsibly in developing this powerful technology. Jacob Cawson, who served as a researcher at both firms, stated on social media that he quit over fears that Anthropic and OpenAI are "gambling with our lives." He said the people building AI "genuinely believe it could kill us all before the end of this decade."

For Ohanian, nuance is critical. "We should talk about the real things that matter, rather than things that might score some quick political points, and I have confidence we can have that conversation." He added, "It's frustrating to me because I see all this misinformation flying around, but I do think in the long run we'll get it right, and that's what I hope happens. So I'm not particularly worried about that part."