Regulators, experts, and AI vendors talk a lot about alignment: the degree to which an AI model conforms to values. But enterprises are still coming to terms with what or whose values should be prioritized.
Credit: Hurst Photo / Shutterstock
In 2003, Oxford University professor Nick Bostrom asked what happens if you ask a smart AI to make as many paperclips as possible. Say, a paperclip manufacturing CEO deployed an AI with no guardrails, checks, or balances, and asked it to apply its superintelligence to the paperclip problem. According to Bostrom, the AI starts transforming first all of earth, and then increasing portions of space, into paperclip manufacturing facilities.
But even an innocuous instruction, like “be helpful,” can cause problems if the AI helpfully assists a hacker to break into corporate systems, or give discounts it’s not supposed to give.
We’ve already seen many examples of corporate AIs not performing as intended. Late last year …