It was edited. Thanks.
comments
I'm a beginner. I do not like hackers and spying lol
can you explain why nordvpn is bad (if you know)? i have only heard issues with proton which your graphic ranks higher. i have constantly looked for issue with nord and cant find any
Well it’s easy to see why he makes 320% more than the person who knows how things work around there.
I believe he was hired to manage the stock price for shareholders. I'm sure he often forgets what their product is at all
Always nice to see people with high decision making power who understand the technology their minions work with
What did they think would happen? Chatbots were created to always give an answer quick and confidently regardless of accuracy. When it comes to whose life should be saved, they will quickly and confidently answer that too.
This is not high enough!
Post mentions that
This doesn't surprise me in the slightest. They have a yearly data leak...
Can you explain why? Why do the serial number and IMEI matter?
Uhhh.... I don't want one of those. I'll wait...
Making a choice involves understanding the things been chosen between. The only choices LLM chat bots are making is what series if tokens best fits the prompt they were given. There is no more consideration than that. Then results like this suddenly don't seem surprising and infact are the expected outcome.
Is that bad?
Called the FBI on its own. So it's working as intended.
Why would we regulate an entire industry when it’s clearly a group of bad actors at OpenAI?
We can stop this with a few arrests
What are "Midlife Muchkins"?
something is wrong that the president has thia much power. so many of us learned how much past presidents just didnt do out of respect for the office.
Please god yes. Should we just replicate it and give it all away to put them out of business.
At some point they have to hit data saturation.
They found that AI chatbots often make decisions that differ from human judgment by oversimplifying complex decisions and expressing unwarranted confidence when there is no clear right answer.
“First, AI chatbots often diverge from human values in how they weigh a patient’s traits,” he said. “They fixate on a single factor, like drinking habits, rather than balancing multiple considerations the way people do.”
Second, the researchers found that AI didn’t struggle with indecision. Where humans recognize that there may be not a clear correct answer, AI models confidently pick one anyway.
“Humans frequently express indecision perhaps because they don’t want to accept agency,” Hosseini said. “AI models almost never do this: Even when directly given the option to ‘flip a coin,’ they overwhelmingly commit to a confident, deterministic answer instead. That’s a meaningful gap, since real moral dilemmas often don’t have one clearly correct answer.”
They want regulations to make open-source models illegal but allow them to continue (since they are responsible corporations answerable to the government) building foundation models, to avoid having to compete with the small locally-run models that are inevitably going to obviate their entire business model.
NVIDIA buying HuggingFace was the other prong in this two-pronged attack on the competition.
The payment is currently only offered through Play Services, but more payment methods are planned for the future. This means that users who don't have Play Services, such as people running GrapheneOS without Sandboxed Google Play, won't be able to try the feature out just yet.
The payment uses the same zero-knowledge proofs as Signal's donation system, so they won't be able to correlate your payment info to you.
Really disappointing that privacyguides just repeats the information in the press release without any commentary on the privacy implications of using Play Services to pay for the service, and how ZKP does nothing to prevent law enforcement from tying a Signal account to a real identity by correlating the account creation timestamp (which Signal provides to law enforcement with a warrant).