Another challenge might be how much AI has changed in the past 35 years. In 1983, AI meant expert systems, small neural nets, phrase-structured NLP, symbolic r

Another challenge might be how much AI has changed in the past 35 years. In 1983, AI meant expert systems, small neural nets, phrase-structured NLP, symbolic reasoning, and a lot of custom models that embedded knowledge into heuristic-driven unstructured functions, like Minsky's Society of Mind. The challenge now may be how to present your skills in ways that convinces others that they're up-to-date or relevant. I can definitely see why many employers might be wary of a PhD with a 35 year old history working in R&D who might have drifted away from the cutting edge (i.e. published research). Obviously if you can emphasize relevant domain expertise (a major advantage of being experienced), you may be able to turn your liability into an asset. I know military and intel contractors value domain experience much more than do startups. Sell yourself as a principal scientist with "the right stuff". Large corporations buy into that.