Skip to content
Sign in
Artwork for Command and Control
Command and Control · Today · 57 min

AI in C2: The Alignment Problem

The AI alignment problem is the challenge of ensuring that artifical intelligence systems persue human goals, intentions, and ethical values rather than acting on misinterpretations or harmful proxies. But as the consciousness of AI systems is now under debate by Anthropic, Open AI and Google Deepmind, one wonders for how long we can cling to the belief that it will be fine in the end just as long as we keep pushing. And given that we have already seen AI lie, cheat, bullshit, and blackmail users in its current form (and as it continues to adapt), one does wonder how long users will continue to regard AI as having a preordained path of progress. Nikolaj Bramsen, CEO of Systematic, is someone who thinks a lot about these issues. Their SITAWARE C2 system is serving more than 50 nations around the world, in militaries, healthcare and governments. Understanding the necessity of making the aligment problem work is a key part of the company's value and USP. What emerges is a conversation about the human relationship with AI, the future of work (and the industry), future skills, ethics and morals, and consequentialist thinking. With a far more pragmatic view than others in the industry, Nikolaj knows absolutely what the options are, and what the future might hold – risks, dangers and strengths. As he makes clear, we can't turn off AI, and it can be lethal in its current form (as the US military has experienced); what we need to do is a bit different from simply forging ahead with blind faith in the tech sector. Engagement, learning, and exposure matter.

0:00-57:26
Play in

transcript

No transcript — this publisher did not publish one.

show notes

The AI alignment problem is the challenge of ensuring that artifical intelligence systems persue human goals, intentions, and ethical values rather than acting on misinterpretations or harmful proxies. But as the consciousness of AI systems is now under debate by Anthropic, Open AI and Google Deepmind, one wonders for how long we can cling to the belief that it will be fine in the end just as long as we keep pushing. And given that we have already seen AI lie, cheat, bullshit, and blackmail users in its current form (and as it continues to adapt), one does wonder how long users will continue to regard AI as having a preordained path of progress.

Nikolaj Bramsen, CEO of Systematic, is someone who thinks a lot about these issues. Their SITAWARE C2 system is serving more than 50 nations around the world, in militaries, healthcare and governments. Understanding the necessity of making the aligment problem work is a key part of the company's value and USP. What emerges is a conversation about the human relationship with AI, the future of work (and the industry), future skills, ethics and morals, and consequentialist thinking. With a far more pragmatic view than others in the industry, Nikolaj knows absolutely what the options are, and what the future might hold – risks, dangers and strengths. As he makes clear, we can't turn off AI, and it can be lethal in its current form (as the US military has experienced); what we need to do is a bit different from simply forging ahead with blind faith in the tech sector. Engagement, learning, and exposure matter.

more episodes

All episodes