Hallucinations, factual errors, a lower in public curiosity, and a plunge in funding —all of those and extra have been talked about lately in information and feedback that appear like the harbinger of a severe setback for Generative AI. Some have even referred to as for the demise of the AI “bubble,” one thing they “predicted a very long time in the past.”
On this submit, I talk about how the present hurdles of Generative AI programs may very well be (have been?) mitigated with the assistance of the great previous symbolic reasoning.
However let’s first clarify these present hurdles.
Take, as an example, one thing as fundamental as arithmetic. I requested yesterday to ChatGPT how a lot is the cubic root of 123,456.78, getting 49.3989 as the reply. Then I requested ChatGPT to multiply 49.3989 by itself two occasions, supposedly getting again to my 123,456.78 quantity, and –no kidding– ChatGPT got here with 123,456.78 consequently, supposedly verifying the correctness of its operation. I may have believed it, however simply to make certain, I made the operation with a calculator and multiplying 49.3989 by itself two occasions acquired 120,545.73, which is nearly 3,000 items away from the supposed outcome. Meaning, in fact, that the primary operation (the cubic root) was unsuitable within the…