Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Sam and Greg's response to OpenAI Safety researcher claims

twitter.com

Post reply on HN
Print view View original

481–490 of 508 posts

Re: Sam and Greg's response to OpenAI Safety researcher claims

#481

Post by bbor »

euroderf wrote: post #478
tempodox wrote: post #458

Earlier quoted context omitted.

> I do believe it's going to replace every single person's job… Some may try to make that so, but since LLMs perform badly with anything that has to be ground in facts and actual reality, the hype wave will die down soon enough. When the point of a job is to generate the most amount of bullshit as quickly as possible, LLMs will excel.

So, marketing as a profession is endangered.

Fingers crossed. Ive come to the slow realization that marketing might be an unjustifiable rot that only makes money through pervasive negative externalities — at least when it’s become so extensive that “brand” advertising is a thing. Would a future of commerce that looks more like eBay / early Amazon than Google really be that sad?

Obviously, this is rough for people who have invested time and money into that profession. Luckily most of them are smart, diligent professionals, so the optimist side of me hopes there will always be a need for such people in our struggles against suffering. For one, we could certainly use a few more teachers…

Re: Sam and Greg's response to OpenAI Safety researcher claims

#482

Post by asadotzler »

_heimdall wrote: post #228
TaylorAlexander wrote: post #193

Earlier quoted context omitted.

> DEI conceptually is a really important idea and could lead to really important change or course correction. As implemented today though, it seems to most often be implemented as a much more surface level program that sure feels more like PR or legal protection than anything else. I suspect that some DEI efforts are helpful and effective, and some DEI efforts are hollow or foolhardy. We probably can’t speak of “DEI…

I do agree that my referring to DEI today may be too broad, that's a great point. > Instances where it works well and gets more people hired and engaged are less interesting to a predominantly white society, so maybe aren’t discussed as much outside of non-white communities. This got me curious, have you sewn any examples of DEI programs helping to get more people hired rather than different people hired? Either can…

Our DEI program was great. It helped us scale from 100 people to 1000 people by scouring HBCU colleges across the US for talent. Had we hired only from Silicon Valley and Stanford, where we were located, things would have sucked during that growth and our previously global hodgepodge of a team that built products bringing in hundreds of millions of dollars would have been White-washed by rich kids, atrophied and died. Instead, we we went on to billions of dollars because we had a group that weren't all fucking Stanford grads with a couple years of Google or Facebook under their belts and mommy and daddy to fall back on.

Re: Sam and Greg's response to OpenAI Safety researcher claims

#483

Post by asadotzler »

behnamoh wrote: post #11
skepticATX wrote: post #2

Interesting that OpenAI has completely destroyed their reputation over the last year. For me, they've gone from an admirable company to cringe-worthy. I really think that the best outcome here is for OpenAI to collapse and get absorbed by Microsoft, where adults can continue some of the good work that is being overshadowed by ego and cult-like vibes.

> cult-like vibes. What cult-like vibes? The OpenAI employees wanted Sam back because of financial reasons and peer pressure. Sam has learnt how to keep employees on leash via money.

What you describe is precisely cult-like and so it gives off cult-like vibes.

Re: Sam and Greg's response to OpenAI Safety researcher claims

#484

Post by bbor »

renegade-otter wrote: post #434
sensanaty wrote: post #430

The talk of AI exterminating us via paperclips is an annoying one to me that detracts from the real dangers that are currently unfolding: - scammers are having the time of their lives, and their capabilities are only going to get better and easier - people are actively getting dumber as they trust the hallucinated outputs blindly - regular communication is starting to break down because you can't trust that people ar…

I sort of cover that in "SkyNet is not coming to kill you": https://renegadeotter.com/2024/04/22/artificial-intelligence... AI can generate a lot of automated output, that is true, but my argument is that it's almost always mediocre at best. There is a LOT to be said about the dangers of AI coming from the people running and funding it - not the AI itself. That stuff is complete bullshit.

Just to voice the opposing view for lurkers: no, the safety concerns about this new and unexpected technology are not unfounded, and dismissing them because chatbots make mistakes is completely missing the point. The point is that experts have been thinking about AI for a long time, and have identified the theoretical limits that prevented us from reaching (‘real’ / ‘general’) AI. Those experts (other thank Chomsky, for now…) are now sounding the alarm — with the advent of intuitive algorithms, the Frame Problem dissolves and a ton of previously intractable problems become tractable.

The point isn’t that chatbots will take over the world through a browser window. The point is that someone’s going to connect 10,000 specialized GPT agents into a series of self-organizing hierarchical agencies, and then hook that up to their country’s justice system, propaganda & surveillance systems, and military systems. Not to mention replacing knowledge workers in mass and the well-discussed potential fallout therein.

We are living in the most exciting, important time since the world wars at the least, and I hope that choice of example drives home the seriousness. Short term, “Solidarity” and “basic human rights” need to enter all of our vocabularies, IMO…

Re: Sam and Greg's response to OpenAI Safety researcher claims

#485

Post by vwkd »

rubyfan wrote: post #446

Something about the use of first names “Sam and Greg” to communicate a friendly relationship with corporations is bothersome to me. It strikes me as phony in its attempt at being casual and human, and at the same time attempting to elevate these people by first name in the popular culture seems pompous.

Yes, I too increasingly get the feeling that OpenAI's leadership wants to become the "next Apple". To me it seems more like a fancy masquerade ball than real solid ingenuity.

Re: Sam and Greg's response to OpenAI Safety researcher claims

#486

Post by lossolo »

sensanaty wrote: post #430

The talk of AI exterminating us via paperclips is an annoying one to me that detracts from the real dangers that are currently unfolding: - scammers are having the time of their lives, and their capabilities are only going to get better and easier - people are actively getting dumber as they trust the hallucinated outputs blindly - regular communication is starting to break down because you can't trust that people ar…

There is one more serious issue: currently, to establish a dictatorship, you need the support of many people because the armed forces consist of people. Now imagine that the armed forces are mostly autonomous robots.

I also agree with your point about virtual girlfriends. Add VR to that, and young boys who are increasingly socially isolated will use this technology to meet their needs.

Re: Sam and Greg's response to OpenAI Safety researcher claims

#487

Post by llamaimperative »

nathanasmith wrote: post #473
llamaimperative wrote: post #387

Earlier quoted context omitted.

Umm, no. We can point to concrete mathematical and empirical evidence of those two things being impossible. Point to the physical laws that indicate AGI or superintelligence is impossible or even maybe-impossible. No such laws exist.

[flagged]

Is this HN-quality commentary?

Re: Sam and Greg's response to OpenAI Safety researcher claims

#488

Post by eli_gottlieb »

sensanaty wrote: post #430

The talk of AI exterminating us via paperclips is an annoying one to me that detracts from the real dangers that are currently unfolding: - scammers are having the time of their lives, and their capabilities are only going to get better and easier - people are actively getting dumber as they trust the hallucinated outputs blindly - regular communication is starting to break down because you can't trust that people ar…

>I don't think AI is going to send out nukes, I do believe it's going to replace every single person's job and make the powerful even more powerful.

The shitty part being: it won't replace people's jobs at equivalent skill and quality levels. It'll just make the output and consumer experience much, much shittier, because for some weird-ass reason a lot of these executive types seem to literally believe that minimizing labor costs is more important than providing value for the customer.

Re: Sam and Greg's response to OpenAI Safety researcher claims

#489

Post by ben_w »

troupo wrote: post #332
m3kw9 wrote: post #100

Earlier quoted context omitted.

Translation: we put in these show “safe guards” to satisfy the law and people peoples nightmare after watching [your AI extinction movie here], but we already had enough safe guards to begin with. They also do not wish to die if you think they have no incentive to keep everyone safe.

There's literally no chance of OpenAI being anywhere close to AGI, and they know it. There's literally no chance of doomsday scenarios, and they know it. However, they do know that AI/ML is used in many applications where mistakes are costly, they don't know how to properly deal with it and don't care as long as they can get people to use it. So they are paying little more than lip service to safety because they know…

What you're saying is what Yann LeCun says (and therefore presumably what he thinks), and appears to be what Facebook is doing by publishing their models (just like they appear to have paid little more than lip service to content moderation leading to them being named by the UN as having played a "determining role" in the Myanmar genocide).

I know people at OpenAI, who have said "we should join the 'Pause AI' protest [outside our office]", so I have reason to think OpenAI is 100% sincere.

They may also be wrong, I hope they are, but definitely 100% sincere.

Re: Sam and Greg's response to OpenAI Safety researcher claims

#490

Post by renegade-otter »

bbor wrote: post #484
renegade-otter wrote: post #434

Earlier quoted context omitted.

I sort of cover that in "SkyNet is not coming to kill you": https://renegadeotter.com/2024/04/22/artificial-intelligence... AI can generate a lot of automated output, that is true, but my argument is that it's almost always mediocre at best. There is a LOT to be said about the dangers of AI coming from the people running and funding it - not the AI itself. That stuff is complete bullshit.

Just to voice the opposing view for lurkers: no, the safety concerns about this new and unexpected technology are not unfounded, and dismissing them because chatbots make mistakes is completely missing the point. The point is that experts have been thinking about AI for a long time, and have identified the theoretical limits that prevented us from reaching (‘real’ / ‘general’) AI. Those experts (other thank Chomsky,…

Well, yes, but the problem is still the people, and these warnings are usually not clear, specific, or just confusing.

Then there is also the media, which breathlessly "translates" what you just wrote into "artificial intelligence is about to go sentient and kill us all".

Post reply on HN