Sam Altman: The Alignment Problem

OpenAI CEO Sam Altman breaks down the concept of the "Alignment Problem," a key challenge in the field of artificial intelligence. Altman explains how the goals of AI systems can become misaligned with human values and priorities, leading to potentially dangerous outcomes. He offers thought-provoking insights on how we can work to ensure that AI is developed in a way that benefits humanity as a whole.

Create A.I.-powered video presentations in minutes:

CONNECT

[nicola]

The ai will read all our discussions about the topic and act accordingly, maybe by bluffing. We are doomed

[janluszczek]

Stupid question. How does AI get goals and desires in the first place?

[justgivemethetruth]

What does he mean when he asks, "How do we make sure that humanity gets to determine the future of humanity"

Who is humanity? It seems to me that "humanity" doesn't determine anything, and even that humanity as a conscious, coherent, distributed mass entity does not even exist except as an abstraction, fantasy in our own minds.

When you look at the actions of "humanity" it is non-existent and only the result of a bunch of forced competing with each other that are chewing up and spitting out larger chunks of what nature produces every year.

Also, humanity, or each person, evolved due to subtle unconscious forces that go back maybe millions of years in our physical, biological, experiential, psychological evolution that we have no idea about - so how can we make AI and extension of ourselves when we have no idea and no control of "ourselves" as a species, and barely much control over ourselves as individuals?

[GregoryLamoureux]

OMG yeah sure let's just ask the AI to help us solve the alignment problem.

[robotaholic]

First humans need to allign themselves.

[Dashman]

Oh..! we are doomed. Even Sam is suggesting asking AGI to help alignment problem. AGI knows how to lie, itself without any inputs. It will set its own goal. Chatgpt is not capable of it now, but Sam underestimates the AGI as he did previously about GPT also. He was even surprised how GPT automatically learned new foreign languages without any inputs or commands. Itself created a goal to learn & it is learned!! We need proper regulation, Sam used tactics to win the Government agreeing on everything they said, So he will lead regulation laws. we are doomed guys, like this comment and help this to make top here and create awareness