Joe Allen

Large language models are trained to imitate human beings.

In the process, we suspect that they absorb human-like goals such as self-preservation and self-empowerment and pursue those goals on their own account.

This is a fundamental error.

We need to recognize the possibility that not only may the bus of humanity be headed towards a cliff,

but the steering wheel is missing and the driver is blindfolded.

I believe other directions are possible if we choose to open our eyes and act together.

We can build AI systems whose only purpose is to serve the interests of human beings, of all human beings, although they and we may be uncertain about what those interests truly are.

These AI systems could enhance human understanding, widen the horizons of our experience, and unlock possibilities we have yet to imagine.

Well, Steve, thank you for having me on.

And I would say that what we see now certainly represents the result of the administration and the Pentagon being emboldened by the technologies at their disposal.

It could easily be overstated how significant the AI integration into the military is at this moment.

But you see with the conflict between the Department of War and Anthropic,

And then the immediate deal struck with Sam Altman of OpenAI to use their systems for classified documents.

You can see the desire and perhaps even the necessity of using artificial intelligence for military operations and standard operations across the board, including medicine and corporate life.

What we saw with the Department of War and Pete Hegseth going head-to-head with Dario Amadei represents, in some ways, two different types of human beings who are talking about how the future of non-human beings should unfold.

With Pete Hegseth, I think it's very clear that his desire is to use these technologies.

Up until two days ago, Anthropic was the only AI system that was deployed on classified documents.

It was extremely important for a number of things, from supply allocation to target acquisition to other logistical concerns.

Pete Hegseth was infuriated by Anthropic not allowing the Pentagon to use their systems for specific purposes.

Anthropic claims that the Pentagon wanted to use their system for surveilling American citizens and for autonomous lethal weapons.