A fascinating and somewhat scary aspect of artificial intelligence is that nobody completely understands how it works. Perhaps for the first time in human history, we have invented technology we do not fully understand.
Many people understand the basics; Machine learning, transformer model, calibration and training, but not why chat bots give the answers they give or why the models show special behaviors.
"People outside the field are often surprised and alarmed to learn that we do not understand how our own AI creations work. They are right to be concerned: this lack of understanding is essentially unprecedented in the history of technology".
– Anthropics vd Dario Amodei
A crucial difference between artificial intelligence and traditional programs is that AI gets its own abilities that appear, like knowledge of mathematics and programming. Common programs have specifically written functions, while AI learns from statistical patterns in data.
Many have pointed out precisely the incomprehensibility of AI, including AI experts and Open AI CEO Sam Altman, who compared the unpredictability of artificial intelligence with the human brain. Most recently, Anthropic Manager Dario Amodei has pointed out that they do not understand their own creations, in a post on their blog.
AI is still moving forward fast and it will be exciting (scary?) To see where we end up in five to ten years.
Recommended Comments
Join the conversation
You are posting as a guest. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.