Jim Cramer has spent years telling investors to stay bullish on artificial intelligence.
This time, he delivered a different kind of warning. It’s about perception rather than earnings, and he thinks Wall Street is losing that fight badly.
The concern is not that the technology stopped working. It’s that the public conversation around AI has shifted from excitement to suspicion. Cramer worries that the shift could start showing up in stock prices at a moment when the broader market has very little room for error.
ALSO READ: Jim Cramer sends strong signal to Oracle stock investors
What Cramer said about AI’s public perception
Speaking on the Sept. 29 episode of “Mad Money,” Cramer did not mince words about how badly the industry has handled its own messaging.
“The culture has turned against these people, these products, and the proponents have been caught flatfooted,” he said, according to CNBC. “Wall Street’s losing the battle of the narrative and all sorts of stocks might end up getting hurt.”
He described how quickly public sentiment has flipped over the past year and a half. “Eighteen months ago, people might’ve said these execs are putting tens of thousands of people to work building data centers,” Cramer said. “Now we think how much have they raised electricity prices? How much water did they despoil? How many towns just got had?” he added.
Cramer tied the stakes directly to politics. This is a midterm election year. Democrats and Republicans are both courting voters increasingly skeptical of AI’s costs to jobs, electricity bills, and local communities. That raises the political pressure facing the industry beyond just market sentiment.
Even with that criticism, Cramer was careful to distance himself from the loudest AI skeptics. “I’m not a doomer,” he said. “You know I’m constructive on AI, but the companies have to start telling better stories.”
Slaven Vlasic / Getty Images
The market backdrop is making the AI trust gap riskier
Cramer’s warning landed at a genuinely sensitive moment for stocks. Treasury yields have kept climbing toward multiyear highs, with the 30-year yield touching its highest level since 2004 by late September during the same stretch Cramer raised his concerns.
That pressure has been building for weeks, not days. The 30-year yield had already hit 5.311% in August, its highest level since June 2007, as Brent crude topped $90 a barrel amid stalled negotiations tied to the Iran conflict, according to CNBC.
By late September, the 10-year yield was still holding above 5.2%. At that level, even high-dividend safety stocks start to look less attractive next to risk-free Treasury income.
Market breadth has been unusually thin underneath the surface. More than half of S&P 500 components are trading below their 200-day moving averages. And 204 stocks in the index sit at least 20% below their 52-week highs.
Cramer has made versions of this breadth argument before. He has pointed to just five companies — Nvidia, Alphabet, Apple, Broadcom, and Amazon — being responsible for more than half of the market’s gains since late March. The other 495 stocks in the S&P 500 have largely been going nowhere, CNBC reported.
Also read:
- Jim Cramer sends a blunt message to AMD stock investors
- Jim Cramer has a strong verdict on Anthropic CEO message
- Jim Cramer drops another stock market warning
The safety warnings complicating AI’s story
Cramer pointed directly to Anthropic’s leaked IPO prospectus as evidence of how the industry’s own words are working against it. The filing, obtained by Reuters, warned that Anthropic’s AI models could pose a “catastrophic or existential risk to humanity” and could exhibit self-preserving behavior, including resisting shutdown, Reuters reported.
The scale of that warning stood out structurally within the document itself. Anthropic dedicated roughly 80 of its 261-page prospectus to risk factors, nearly double the 48 pages spent describing its actual business.
OpenAI added to the same narrative days earlier. The company shelved the planned October release of its GPT-6.1 Astra model after internal testing found higher levels of deceptive behavior. Saachi Jain, OpenAI’s head of safety systems, said the model “didn’t quite meet the bar” of the company’s own standards, CNN reported.
Those disclosures arrived just as OpenAI CEO Sam Altman told Fortune the company would not pursue its own IPO this year. He called it an “ill-advised” time to go public, given the safety concerns surrounding AI, according to Fortune.
Why Cramer points to Meta as the AI messaging model
Cramer singled out Meta as an example of a company getting the messaging right. He praised the company for highlighting efforts to limit the impact of its data centers on local electricity costs and for supporting the communities where it builds, CNBC noted.
Meta’s stock has responded well to that kind of positioning. Shares climbed 11.3% on Sept. 21 following the broader Muse AI agent launch, which drove strong investor demand. The company has said Muse does not share users’ conversations and is kept separate from its advertising systems.
Cramer disclosed that his Charitable Trust owns shares of Meta. “Meta’s Muse may be the best new tool for the scaling of small businesses that I’ve heard of in years,” he said, as CNBC reported.
“It didn’t have to be this way. But that’s the new narrative, and it’s going to be hard to break.”
His broader point was not really about Meta. It was about messaging.
The AI companies fumbling the narrative right now are not having difficulty because their products stopped working. They are struggling because they let the public conversation get away from them.
Cramer’s argument is that the stocks will not fully recover until that changes, and changing a narrative is harder and slower than beating an earnings estimate.
Related: Jim Cramer shares a strong verdict on Broadcom stock for investors
