AI and Hitting a Math Wall
March 25, 2026
Another dinobaby post. No AI unless it is an image. This dinobaby is not Grandma Moses, just Grandpa Arnold.
The average AI chatbot user realizes that the technology has its limits. An intelligent user (who doubles checks their facts) knows that the bots are prone to hallucinations and takes everything it dishes out with a binary salt grain. Gizmodo explains the limits of AI bots and how the technology is about to hit a computational brick wall: “AI Agents Are Poised to Hit A Mathematical Wall, Study Finds."
AI bots are built on LLMS with the belief that they will infinitely grow, gain more knowledge, and become more human in their autonomy. The father and son research team, Vishal Sikka and Varin Sikka, wrote a paper (hopefully without AI’s help) about the limits of AI. Apparently LLMs can’t do agentic and computation tasks beyond a certain complexity. In other words, AI may face computational limits. Thus, mathy innovation is going to be needed.
The paper explains that AI are programmed to complete tasks only as far as the parameters of the LLM. LLMs have limited processing capabilities and must operate within its their bands of knowledge. When tasks go beyond those parameters, more complex models are needed. The LLMs can’t extrapolate the required information so they either fail in the tasks or return incorrect information.
AI, therefore, needs to be helped out with humans who come up with new methods and techniques:
"The basic premise of the research really pours some cold water on the idea that agentic AI, models that are able to be given multi-step tasks that are completed completely autonomously without human supervision, will be the vehicle for achieving artificial general intelligence. That’s not to say that the technology doesn’t have a function or won’t improve, but it does place a much lower ceiling on what is possible than what AI companies would like to acknowledge when giving a “sky is the limit” pitch.”
Other experts have reported similar results and the average user can tell you the same thing. Can AI replace humans. No, but the MBAs and bean counters have calculated that smart software is cheaper and faster. Plus, AI does not need health care, retirement contributions, or vacations.
Whitney Grace, March 25, 2026
Comments
Got something to say?
-
- Subscribe to Beyond Search
Feature archive
News archive
-
