Mark Cuban Asked Grok What Will Cause ‘Human Extinction’ in 25 Years — It Says Al Is 3,000x More Likely Than Climate Change

Billionaire investor Mark Cuban put an unusually grim question to Grok on Sept. 14 in a post on X. If humanity faces an extinction event within the next 25 years, is artificial intelligence (AI) or climate change more likely to cause it?

Grok’s answer was artificial intelligence.

“AI is more likely than climate change,” the chatbot replied after Cuban asked it to assign a probability to each scenario and estimate when either event could occur.

Grok put the probability of AI causing human extinction within the next 25 years at roughly 3%. It gave climate change a probability of about 0.001%.

The math is straightforward. Grok’s estimated probability for AI was 3,000 times higher than its estimate for climate change.

Sponsored Content: The Flying Car Market Could Reach $9 Trillion — Here’s One Way to Invest Early

The chatbot said an AI extinction event could occur in the mid-2030s to 2040s through “rapid misaligned superintelligence,” while stressing that the estimate was “highly uncertain and contested among experts.”

For climate change, Grok said there was “no plausible path to extinction that soon.” It said climate change could lead to severe impacts and species loss but argued that humans could adapt through technology and migration. 

The figures need some context. Grok isn’t an authority that can establish the probability of an extinction event, and there is no historical dataset that can produce a precise probability for a future technology reaching a level of intelligence that doesn’t yet exist.

Still, the question touches on a real debate among AI researchers.

AI Researchers Have Put Their Own Numbers on the Risk

The 2023 Expert Survey on Progress in AI, conducted by AI Impacts, collected responses from 2,778 AI researchers who had published at six major machine-learning and AI conferences.

When asked to assume that high-level machine intelligence (HLMI) would eventually exist, respondents gave a median 5% probability to an outcome that was “extremely bad,” including human extinction. The survey also found that 68% thought the long-term impact of such systems would be positive rather than negative. 

That isn’t a consensus that AI will destroy humanity. The estimates varied significantly, and the survey itself was designed to measure how AI researchers view several possible futures rather than establish a single forecast.

It does show why the subject keeps coming up as AI systems become more capable.

AI Investment Keeps Growing

The money flowing into the industry hasn’t slowed because of those concerns.

Global corporate AI investment reached $581.7 billion in 2025, up 130% year-over-year (YOY), according to Stanford University’s 2026 Artificial Intelligence Index. Private AI investment reached $344.7 billion, an increase of 127.5%. The United States accounted for $285.9 billion in investment. 

That capital isn’t going solely toward companies developing consumer chatbots. The spending runs through the wider AI supply chain, including semiconductors, data centers, cloud computing, energy, and software.

The rapid expansion is also creating a market for companies working on AI security, testing and monitoring. As businesses deploy increasingly capable models, controlling how those systems operate becomes part of the cost of doing business.

For investors, that’s where the conversation around AI risk becomes more relevant.

Grok’s 3% estimate doesn’t establish whether an AI company is overvalued, and it isn’t a reason by itself to buy or sell a stock. It does highlight a tension running through the industry. Companies are spending hundreds of billions of dollars to make AI systems more capable while researchers are still debating what happens when those systems become significantly more powerful.

The commercial opportunity is tied to that development. So are some of the industry’s biggest uncertainties.

Cuban’s question put a number on one possible outcome. Whether Grok’s number means anything beyond the conversation that produced it remains an open question.

On the date of publication, Caleb Naysmith did not have (either directly or indirectly) positions in any of the securities mentioned in this article. All information and data in this article is solely for informational purposes. For more information please view the Barchart Disclosure Policy here.