PhilipTheBucket
@PhilipTheBucket@ponder.cat
- Comment on RedNote may wall off “TikTok refugees” to prevent US influence on Chinese users 2 days ago:
That’s actually a really good idea. And LLMs are excellent at machine translation, it’s one of their underutilized killer apps. They’re a little better at going to English than from English or between two non-English languages, just because so much of the corpus is English, but they’re easily enough that it could be automated as a working solution.
- Submitted 3 days ago to technology@lemmy.zip | 33 comments
- Comment on TikTok Users Gleefully Embrace Even More Chinese App To Spite US TikTok Ban 5 days ago:
They don’t mind that it’s ruining people’s brains. They just don’t like that it’s sending data back to the CCP, giving them direct access to a really huge number of American’s phones to do whatever they like, and giving them an uncomfortably free hand in guiding the perception of reality experienced by millions. But the harm it does to people on a personal level is as unimportant to the US government as it is to China’s.
- Submitted 5 days ago to technology@lemmy.zip | 0 comments
- Submitted 5 days ago to technology@lemmy.zip | 17 comments
- AT&T Sets A Comically Narrow Definition Of “Service Outage” After Particularly Embarrassing Wireless And 911 Outagewww.techdirt.com ↗Submitted 1 week ago to technology@lemmy.zip | 0 comments
- Submitted 2 weeks ago to technology@lemmy.zip | 2 comments
- Copyright Industry Wants To Apply Automated Blocking To The Internet’s Core Routerswww.techdirt.com ↗Submitted 3 weeks ago to technology@lemmy.zip | 5 comments
- Comment on Encyclopedia Britannica Is Now an AI Company 3 weeks ago:
I read some of the author’s other articles. They have a habit of regurgitating highly suspect claims from press releases or company self-descriptions as if they were reality.
OpenAI is confident in o3, and offers impressive benchmarks—it says that in a Codeforcing testing, which measures coding ability, o3 got a score of 2727. For context, a score of 2400 would put an engineer in the 99th percentile of programmers. It gets a score of 96.7% on the 2024 American Invitational Mathematics Exam, missing just one question.
There’s also the article which claims that AI puts the entire power grid at risk, and then when you read the article, you learn that in order for that to be true, you need to lump AI in with crypto mining, other datacenter expansion, electric cars, and climate control for people’s homes.
- Comment on Encyclopedia Britannica Is Now an AI Company 3 weeks ago:
More general-purpose models like ChatGPT suffer from hallucinations because they have hoovered up the entire internet, including all the junk and misinformation.
Incorrect. ChatGPT hallucinates because that’s how LLMs work. Hoovering up misinformation is a separate problem.
A company in the space of selling educational books that has seen its fortunes go the opposite direction is Chegg. The company has seen its stock price plummet almost in lock-step with the rise of OpenAI’s ChatGPT, as students canceled their subscriptions to its online knowledge platform.
Incorrect. Chegg is a cheating platform. It is the opposite of a knowledge platform.
Why is Gizmodo paying people to write articles who apparently know pretty much nothing about the subject they are writing about?
- Government to Name ‘Key Witness’ Who Provided FBI With Backdoored Encrypted Chat App Anomwww.404media.co ↗Submitted 3 weeks ago to technology@lemmy.zip | 0 comments
- Submitted 3 weeks ago to technology@lemmy.zip | 3 comments
- Submitted 3 weeks ago to technology@lemmy.zip | 0 comments
- Supreme Court Won’t Help Big Telecom Kill NY Law Requiring Affordable Broadband For Poor Peoplewww.techdirt.com ↗Submitted 4 weeks ago to technology@lemmy.zip | 1 comment
- Submitted 4 weeks ago to technology@lemmy.zip | 0 comments
- Submitted 4 weeks ago to technology@lemmy.zip | 9 comments
- Comment on Study reveals AI chatbots can detect race, but racial bias reduces response empathy 4 weeks ago:
To accomplish this, researchers asked two licensed clinical psychologists to evaluate 50 randomly sampled Reddit posts seeking mental health support, pairing each post with either a Redditor’s real response or a GPT-4 generated response. Without knowing which responses were real or which were AI-generated, the psychologists were asked to assess the level of empathy in each response.
I am doubtful that this framework is going to accurately detect anything at all about the usefulness of chatbots in this context, whether about race or anything else.
I don’t think using chatbots for psychology is a good idea, but this study isn’t the way to study and make that determination.
- Submitted 4 weeks ago to technology@lemmy.zip | 4 comments
- Submitted 5 weeks ago to technology@lemmy.zip | 1 comment
- Submitted 5 weeks ago to technology@lemmy.zip | 3 comments
- Submitted 5 weeks ago to technology@lemmy.zip | 0 comments
- Submitted 5 weeks ago to technology@lemmy.zip | 0 comments
- Submitted 5 weeks ago to technology@lemmy.zip | 6 comments
- Submitted 5 weeks ago to technology@lemmy.zip | 0 comments
- Submitted 5 weeks ago to technology@lemmy.zip | 0 comments
- Submitted 5 weeks ago to technology@lemmy.zip | 1 comment
- Submitted 5 weeks ago to technology@lemmy.zip | 3 comments
- Submitted 5 weeks ago to technology@lemmy.zip | 6 comments
- FBI Official Reluctantly Touts Encryption Since US Telecom Providers Are Still Compromised By Chinese Hackerswww.techdirt.com ↗Submitted 5 weeks ago to technology@lemmy.zip | 0 comments
- Submitted 5 weeks ago to technology@lemmy.zip | 0 comments