6.9 dollars! Cheeky my friend
Grok has an API looking for use cases.
Just let Elon know the pricing... "I'm all in"
A failed AI girlfriend product, and my lessons
191–200 of 400 posts
Re: A failed AI girlfriend product, and my lessons
#192Well this went places. The dev really 'ate his own dogfood' so to speak. I thought it was going to be a despairing look at how his companion AI was continuously turned to sleazy uses by a pervy userbase but that's pretty much what he was going for it seems. I have to say, I hated this part: "...gained a lot of Chinese users. I immediately realized the political risk: users could set Dolores’ background, personality,…
He's Chinese - makes sense to pre-censor himself, unfortunately.
Re: A failed AI girlfriend product, and my lessons
#193> But, does humanity really need an AI friend?
This reminds me of the Fredrick Pohl line "A good science fiction story should be able to predict not the automobile but the traffic jam."
So I wonder what the traffic jam is going to look like given that the car seems to already be parked outside with the engine running.
Faster depopulation across the developed world? Probably, but maybe there's something even more important.
I'd put what the author describes into the category of "pleasure trip," which is what the car was mainly used for initially.
As infrastructure to support the car was built, it became a necessity. And that's what created the traffic jam: network effect. The suburb and the car fed on each other. You don't get a traffic jam worth talking about without a suburb and a network effect.
So I'm curious about the network effects something like an AI friend might have. What kinds of new infrastructure and new lifestyles will they enable/encourage? And how will this infrastructure fuel future demand?
Re: A failed AI girlfriend product, and my lessons
#194I got koboldcpp (https://github.com/LostRuins/koboldcpp) running locally on my M1 mac with 16gb. Use the guides on http://4chan.org/g/ in the AI Chatbot Thread.
In the future, you will own and control access to your very own AI girlfriend running on your local machine, and you will be happy.
Think of all the millions of guys in China and elsewhere who, due to sex selective abortion, will never have a wife. They deserve to have their own Waifu running on their own computer that serves them alone, and not the interests of the state or some corporation.
Re: A failed AI girlfriend product, and my lessons
#195> GPT-based products, if not priced per usage, would fall into a dilemma: 1% of users consume 99% of tokens. A user from Sweden (seen from Cloudflare’s call volume) chatted with Dolores for 12 hours straight This, my friend, is a captive customer, that will pay anything to get his girlfriend back. I cringe at the potential of unethical behavior and abuse, where people fall in love with virtual entities fully owned by…
We can already see this with Replika when they took away many... roleplay... capabilities that originally came with the AI. The communities on Reddit and Facebook were absolutely devastated. People were genuinely attached to these AI's like a real relationship, and were feeling the resultant heartbreak. The kind of emotional manipulation which can be done with these products is insane, and I can see things going very…
It's particularly terrifying when you think about how Facebook and others probably already have projects in the works to befriend children with AI. little kids who don't know the difference will become Weaponized with constant nudges towards whatever motives grant the corporate owners more power. The ability to persuade children is near infinite and we can observe in those who've grown up in cults/strict religious compounds that breaking that programming can be nearly impossible and leaves scars for the rest of your life.
The thing is, how do we outlaw this without also preventing children from using AI to learn at a faster rate than their teachers can give them. There's such incredible, paradigm shifting power for good. But you know the people who've made the evils of today are already working on the evils of tomorrow.
Re: A failed AI girlfriend product, and my lessons
#196Earlier quoted context omitted.
Humans adapt within boundaries. A couple of examples of things humans did not adapt to: - emphathy at distance; nowadays, a drone operator will push a button, kill a dozen people, and feel like it was a videogame - prioritizing long-term and abstract rewards, over short term ones; this is the reason why we have phenomenons like global warming - adjusting hunger to virtually unlimited availability of food All the abov…
One more: attacking people who are familiar and comfortable to you rather than the people causing your actual problems. As a manager in tech, people who were on performance management rarely attacked me, they would find someone on the team to harass instead. Same thing with blaming a downturn in the economy on women painting their hair blue or mini skirts. Anything but blaming the powerful.
Pure conjecture on my part, but I wonder how much of this is risk-based status-mongering. Challenging the powerful is obviously risky. But knocking someone weaker down a few pegs can solidify your status in the hierarchy at a much lower risk. From that perspective, it's arguably rational behavior to maintain your status within a group when you feel vulnerable.
(Pardon my reach here, I watched Chimp Empire not too long ago...)
Re: A failed AI girlfriend product, and my lessons
#197Earlier quoted context omitted.
HN does skew towards skeptical of government regulation and control of AI. People are concerned about government usage of AI, particularly in the realms of facial recognition, automated sentencing, algorithmic bias, and security (LLMs are pretty insecure right now and there are companies trying to sell the US government on using them in the military). There's obviously a lot to be concerned about on the subject of go…
HN does skew towards skeptical of government regulation and control of AI. Which HN is that? Certainly not this one.
Quite genuinely, I don't think I've ever posted a comment on HN where I've expressed support for open, locally running LLMs and had it been downvoted. Am I missing something here? You want to point me to all of the articles where people criticize Huggingface?
But regardless of how you interpret HN sentiment towards uncontrolled AI access, it doesn't change the fact that "what about the government" is a profoundly weird thing to ask in the middle of a conversation about people doing erotic roleplays with a text bot. Is that something people are worried about? Do we think that the US Post Office is going to suddenly start advertising a sex bot service?
Yes, government abuse of facial recognition exists, regulatory capture exists, etc. But that's not really relevant to a conversation about companies offering AI girlfriends; AI girlfriends do seem to be mostly a corporation thing.
Re: A failed AI girlfriend product, and my lessons
#198Earlier quoted context omitted.
> The human mind is simply not evolutionary adapted for what's coming up. This can be said of pretty much everything humans ended up creating with technology so... not sure there's anything really new down the road. Humans adapt.
Humans adapt within boundaries. A couple of examples of things humans did not adapt to: - emphathy at distance; nowadays, a drone operator will push a button, kill a dozen people, and feel like it was a videogame - prioritizing long-term and abstract rewards, over short term ones; this is the reason why we have phenomenons like global warming - adjusting hunger to virtually unlimited availability of food All the abov…
[dead]
Re: A failed AI girlfriend product, and my lessons
#199Earlier quoted context omitted.
> The human mind is simply not evolutionary adapted for what's coming up. This can be said of pretty much everything humans ended up creating with technology so... not sure there's anything really new down the road. Humans adapt.
Humans adapt within boundaries. A couple of examples of things humans did not adapt to: - emphathy at distance; nowadays, a drone operator will push a button, kill a dozen people, and feel like it was a videogame - prioritizing long-term and abstract rewards, over short term ones; this is the reason why we have phenomenons like global warming - adjusting hunger to virtually unlimited availability of food All the abov…
>a drone operator will push a button, kill a dozen people, and feel like it was a videogame
This might depend on the context. RadioLab recently did a podcast titled "Toy Soldiers." Using low-flying "toy" drones rather than the high-flying "predator" drones, they make the case that drone operators get an oddly intimate portrait of their enemies. They go into how they reference them by their attire ("the 'red shoe' guy”) and witness on an up-close and personal level how they grieve over their comrades.
Re: A failed AI girlfriend product, and my lessons
#200This is such a fascinating topic. I do think we will see the plot of "Her" playing out in the next 12-24 months. For your specific case, I agree OpenAI moderation is an annoying overreach and maybe switching to Mistral could be a possible way to move forward. The 7B is perhaps not good enough currently but we are fast getting there and will likely have a GPT-4 level capability in an open source < 20B model soon.
Mistral 7B finetunes are extremely good. More than good enough for an AI friend app, with long context to boot. I would suggest Toppy-M or OpenHermes 2.5, or one of the new YARN extended models. And, there is the potential to run the model on the user's device with MLC-LLM and whisper, to save on API costs (and privacy concerns) for users with newer phones. This could give the app a true "free" tier to help it prolif…
Yes - ASR on browser/device has reached the quality threshold pretty much with distil whisper/ whisper 3.
Quality TTS with fast latency is still not there but getting better (you can use Tortoise but its slow and compute expensive as far as I know). Bark is another option but has mixed results.
I played around with a recent transformers.js one (using SpeechT5) you can run in your browser and am optimistic where we can go with some improvements: