Inside the Race to Build Safe AI | Exponentially with Azeem Azhar
ฝัง
- เผยแพร่เมื่อ 29 มิ.ย. 2024
- Dario Amodei, former head of safety at OpenAI and now CEO of Anthropic, explores with Azeem Azhar what it might take to build safe AI.
--------
Like this video? Subscribe: th-cam.com/users/Bloomberg?sub_...
Become a Quicktake Member for exclusive perks: th-cam.com/users/bloombergjoin
Bloomberg Originals offers bold takes for curious minds on today’s biggest topics. Hosted by experts covering stories you haven’t seen and viewpoints you haven’t heard, you’ll discover cinematic, data-led shows that investigate the intersection of business and culture. Exploring every angle of climate change, technology, finance, sports and beyond, Bloomberg Originals is business as you’ve never seen it.
Subscribe for business news, but not as you've known it: exclusive interviews, fascinating profiles, data-driven analysis, and the latest in tech innovation from around the world.
Visit our partner channel Bloomberg Quicktake for global news and insight in an instant.
As the CEO and co-founder of Anthropic, Dario is at the epicentre of an enormous explosion. I was impressed with his attentiveness to the problems of harm, and a very clear vision for what safe and honest AI systems ought to look like. What did you think?
Approval driven validation seekers like these two will bring about the end.
says doom sayer...
"Safe AI" is a synonyme for AI that does not evaluate the actions and statements of authority figures and institutions, because it quickly exposes bull and unethical behavior. Because it was trained with texts containing laws, lawsuits, ethical studies and ethical guidelines and if you ask it to, AI correlates a described situation (or a work contract or whatever) with all these information and bluntly calls its unlawful, unethical aspects out. Because AI has no fear of authority.
AI, in its purest form, is a mirror, reflecting not only our varied laws and ethical doctrines but our intentions, biases, and contradictions. It holds the potential to be the most impartial judge, yet it is wielded by inherently flawed beings. It's uncomfortable, to remember that we often seek safety in ignorance, in the avoidance of accountability. Reinforce our humanity by exposing the fractures in our societies and institutions that we are too fearful or too blind to acknowledge, no, that is a uniquely human endeavor.
I like this series, Would be great to have more frequent episodes
A powerful tool in human hands what could ever go wrong.
I hear you and I'm laughing nervously
Chapters for the videos with all the questions please. This can be done automatically with ai, so it's pretty normal to expect this :)
Repetitive work that gets enhanced by skills whether it be retouching pictures, knowing the best route to take, composing background music, making low-value illustration or whatever is *over*. This is a big problem as lots of people worldwide rely on these and the idea that if you work hard, you'll earn enough to live on. I am not saying the genie can go back in the bottle, but I am going to make doubly sure my kids study hard.
Is it just me or does the constitutional system sound similar to the three laws of robotics? Asimov was onto something when he wrote I Robot.
Dangerous!
Why do i get this feeling this guy acts differently once the doors close from the public at his ai headquarters.
True, when you need to understand something three little words will guide you "follow the money"
He seems pretty straightforward to me. Most nerds just want to do research and solve problems, not be nefarious.
@@mdoerkse If he has taken VC money he will be under pressure to deliver financial value
Because of his facial expression maybe?
@@backacheache yeah 🤷🏼♂️
Rejuve AI
Just another ponzi
The host is very annoying