4 ms·

The length of the democratization cycle we're seeing – months to weeks between a breakthrough model and a competent open-source alternative that runs on commodi

The length of the democratization cycle we're seeing – months to weeks between a breakthrough model and a competent open-source alternative that runs on commodity hardware – really highlights the genie-stuffing posture of Google and OpenAI. All the thoughtful, if highly paternalistic guardrails they build in amount to little more than fig leaves over the possible applications they intend to close off. I'm personally in the "AI risk is overstated" camp. But if I'm wrong, all the top-down AI safety in the world is going to be meaningless in the face of a global network of researchers, enthusiasts, and tinkerers.

hedora 4y ago
They claim the guardrails are for public good, but they're pretty clearly using them to try to establish a competitive moat. It's similar to the "we don't sell personal information" claim. Sure, but that's because they make money renting malicious actors access to a black box that contains your personal information. Selling the contents of the box would reduce their overall revenue.
TulliusCicero 4y ago
To me it seems like an obvious case of reputational risk being much larger for more prominent organizations than for smaller ones. It makes sense for Google to wait for some startup to "go first" in releasing a model largely without controls. That way, some random startup takes the initial heat of "people are using AI for bad things!!" headlines plastering tech blogs. Then Google can do basically the same thing a little bit later, and any attack pieces will sound old hat.
narrator 4y ago
I was using AI Dungeon with the full power GPT-3 model before they crippled it. That thing had a very uninhibited mind for erotica! Imagine what would happen when that power comes to image models!
- lodovic 4y ago
  It's only a matter of time before someone trains a model on stills from the major adult sites. Actually surprised it hasn't been done / made public yet.
nonbirithm 4y ago
I wonder how it would be possible to stop people from publishing and spreading research that will doom some subsegment of humanity/culture. If it turns out that proliferating something like DALL-E 5 causes serious irreversible effects to human culture, what would the researchers conclude was the correct thing to do at this exact point in time? Stop publishing AI research? How would we get all 8 billion people on Earth to agree? It's the reason why the rapid amount of progress in this field scares me at times. It feels like gradually being crushed under a wall of the inevitable march towards progress. It could be the case that stopping ourselves before it's too late isn't possible. Sometimes I get the feeling that the laws of nature will eventually destroy or severely impact any lifeform that gains too much of an understanding about the world. For example, in some other universe it might be possible to survive X more decades if the laws of physics weakened the effects of nuclear war just enough for civilization to recover in a relatively short period of time, but we're stuck with what we have, and that doesn't necessarily map to the long-term survival of a highly intelligent lifeform. It would be a shame if unbounded curiosity would be humanity's undoing. That curiosity is also a part of me, and my family, and my neighbors down the street, and those in the situation room.
- Aeolun 4y ago
  > How would we get all 8 billion people on Earth to agree? You don’t, so the question is mostly academic.

hedora 4y ago

They claim the guardrails are for public good, but they're pretty clearly using them to try to establish a competitive moat. It's similar to the "we don't sell personal information" claim. Sure, but that's because they make money renting malicious actors access to a black box that contains your personal information. Selling the contents of the box would reduce their overall revenue.

TulliusCicero 4y ago

To me it seems like an obvious case of reputational risk being much larger for more prominent organizations than for smaller ones. It makes sense for Google to wait for some startup to "go first" in releasing a model largely without controls. That way, some random startup takes the initial heat of "people are using AI for bad things!!" headlines plastering tech blogs. Then Google can do basically the same thing a little bit later, and any attack pieces will sound old hat.

narrator 4y ago

I was using AI Dungeon with the full power GPT-3 model before they crippled it. That thing had a very uninhibited mind for erotica! Imagine what would happen when that power comes to image models!

lodovic 4y ago
It's only a matter of time before someone trains a model on stills from the major adult sites. Actually surprised it hasn't been done / made public yet.

nonbirithm 4y ago

I wonder how it would be possible to stop people from publishing and spreading research that will doom some subsegment of humanity/culture. If it turns out that proliferating something like DALL-E 5 causes serious irreversible effects to human culture, what would the researchers conclude was the correct thing to do at this exact point in time? Stop publishing AI research? How would we get all 8 billion people on Earth to agree? It's the reason why the rapid amount of progress in this field scares me at times. It feels like gradually being crushed under a wall of the inevitable march towards progress. It could be the case that stopping ourselves before it's too late isn't possible. Sometimes I get the feeling that the laws of nature will eventually destroy or severely impact any lifeform that gains too much of an understanding about the world. For example, in some other universe it might be possible to survive X more decades if the laws of physics weakened the effects of nuclear war just enough for civilization to recover in a relatively short period of time, but we're stuck with what we have, and that doesn't necessarily map to the long-term survival of a highly intelligent lifeform. It would be a shame if unbounded curiosity would be humanity's undoing. That curiosity is also a part of me, and my family, and my neighbors down the street, and those in the situation room.

Aeolun 4y ago
> How would we get all 8 billion people on Earth to agree? You don’t, so the question is mostly academic.