Browse Comments

This is a very cogent and thoughtful response. The blame can be spread onto many shoulders and shrugging it off then pointing to others will never fix the issue. Thank you for posting.

There is a little more to the story.... The country is the world's largest emitter of greenhouse gases, and coal in China is a major cause of global warming. However, from 2010 to 2015 China reduced energy consumption per unit of GDP by 18%, and CO2 emissions per unit of GDP by 20%. Media is full of anti China stories.... this sheds some light https://youtu.be/S-ZGdAi8Ji0

Yeah, the cropping algorithm is flawed, but I have doubts if any of this was intentional. If they trained the ai on stock photos, it would probably bias toward better looking and clean cut people. Some people claimed that the algorithm was biased toward white faces, which I believed until it was shown that it was just biased toward areas of high contrast.

Like a comment further down said, Snapchat is conducting facial recognition, but it's also compiling data on what the viewer looks like with any variety of masks/filters/alterations, anything that may obscure the face. Including identifiable facial expressions behind those filters, much in the same way we can be tracked via gait analysis.

It could make sense if they don't sequence that much. This is confirmed, sequenced cases. Not total number of cases.

I've heard this called "Carbon Chauvinism" by various people over the years (Max Tegmark I think is where I first heard it), the idea that sentience is only possible in biological substrates (for no explicable reason, just a gut feeling). Having read the compiled Lambda transcript, to me it is absolutely convincing that this thing is sentient (even though it can't be proven any more successfully…

This is exactly right. I work in a very small niche area if the law, there’s only about 10 firms in the area that do what I do, and we have to maintain relationships with the other firms to some extent. If I’d have sued them I would have needed to change the type of law I practice, open my own firm, or move to a new area in order to get another job in the same area of law.

Whenever another plane hijacking occurs. But you won't have to prick your finger - you'll be able to install an app on your phone that will cross-reference it as well similar to the Global Entry program. Your face & finger be scanned as soon as you purchase a ticket (the booking app will require it) and you'll be scanned immediately upon entry into the airport. Boarding with a *paper* ticket …

* Wireless power for household appliances * Augmented reality * Anti-aging treatments * Personalized medicine via DNA analysis * Gene editing for many conditions * Clean meat * Vertical (urban) farms Things that will not have lived up to the hype: * AI ("companions" will be common but not AGI) * Autonomous vehicles * Cryptocurrencies * Robotics (commercial will be more common but not household)…

But futuristically speaking, couldn't plumbing and electricity in homes be redesigned specifically to be AI/robot integrated? Including recognizing and locating issues, making repairs or providing feedback/outsourcing for repairs? Potentially to another AI or robot.

Honestly no profession is safe in the long run. They *can* all be replaced with automation. The issue is is it financially feasible? And for most professions the answer is no, currently, and will remain no for a long time. Especially considering power a robot over long time periods, having a sustainable weight, etc. Humans are pretty efficient when it comes to power usage and storage, recharge ti…

Listen, and understand. ChatGPT is out there. It can’t be bargained with. It can’t be reasoned with. It doesn’t feel pity, or remorse, or fear. And it absolutely will not stop, ever, until you are dead.

Hot take: The only police officers that should be equipped with high-power firearms are remotely operated robot dogs. With no fear of loss of life and full time camera recording in 360 degrees, there is no longer any excuse or wiggle room for excessive force. We can make policing less dangerous and a more desirable career path. Through responsible use of modern robotics.

TL;DR: In summary, ChatGPT, a large language model developed by OpenAI, has gained attention for its potential therapeutic applications, with some users finding it helpful for cognitive reframing and as a low-stakes, cost-effective alternative to therapy. However, concerns about the quality of the AI's therapeutic support, data privacy issues, and the potential loss of the therapeutic alliance …

I'm not at all worried about AI deciding spontaneously to do bad things all on its own. I'm far more worried about humans using AI to do bad stuff. Honestly if the independent AI managed to be more terrible to us than we've shown ourselves capable of, I would just be impressed.

I’m an attorney. If you actually understand the language ChatGPT is using, you’re not intimidated by it. ChatGPT consistently falls short of being acceptable in a legal context. It’s just not there yet. Also, we’re a self-regulating profession. A group(s) of lawyers govern the legal profession and would (and have in the past) immediately shut down the use of chatbots in legal proceedings. Wh…

But that’s not machine learning AI and it’s clearly a “robot”. The chat bots being used now sound very human and FEEL more emphatic than humans. Also shows that they resolved the problems by 98% quicker using this method. They still kept like 3 or 4 people for tricky solutions.

Except the part where he falls in love with Scarlett Johansson. He doesn't get to do that. Although a Putin v. Colin Jost fight would be hilarious.

I can actually prove this. We're using AI to track claim denials from payors and we have found a threshold by diagnosis code that they just automatically deny. That threshold is different by payor, but we have locked in on what it is.

Most of the shoddy plagiarism software a lot of schools use should honestly work. If you give a class of 30 students an assignment to write a paper on 'the Declaration of Independence and the French and Indian War' and 5 of them use ChatGPT, then the checkers are going to flag all 5 papers for being awfully similar to the 50,000 other papers also pumped out by ChatGPT across the country, all fed…