David Brochu: AI will not kill society; we’re doing just fine by ourselves

To The Daily Sun,

Recent very public disclosures of significant misbehavior by rogue agents has only lately caught the attention of our hapless political elites. The disingenuous hand wringing and calls for a pause by the heads of both OpenAI and Anthropic, as well as a general outcry that AI has a non-zero chance of being an extinction event and out to be stopped, is naked attempt to deflect the truth.

We passed the point of no return a long time ago and everyone building this new thing knows it. Current abilities of even open weight models are more than sufficient to engineer any number of species ending catastrophes and has been for sometime.

They've know this for sometime now. Recent declarations not withstanding. No one is stopping nor should we.

Bad actors will use AI to achieve their aims. Like it or not we are in an AI arms race. What is at risk is not extinction we see that that risk. Note that no AI has had its own terminal objective. Every single "rogue" actor was acting in service of its master goals. This is critical. AI amplifies our ability to destroy each other. It is a known risk. Known risk can be planned for.

The real risk is human irrelevance. The car did not kill off the horse. It turned it into a pet.

Embodied recursively self improving AI aka Robots, will be our superior in every way.

The only way forward is to align AI with universal non-fungible human values. History would indicate it can be done.

History also points to humans learning the hard way. Let's hope we can change that pattern while we still have time.

David Brochu

Belmont