492 karma · joined May 31, 2022
https://soundcloud.com/wort-fm/christine-wenc-on-the-legacy-...
https://soundcloud.com/hachetteaudio/funny-because-its-true-...
https://www.nytimes.com/2025/03/12/books/new-nonfiction-book...
I'm a strong advocate for a united effort to create a training set of the collected works of mankind free to any AI company to use if, for instance, it uses its profits to fund UBI, or some other program to pay us for what they use.
Anything but AGI profits being used to keep score in the Oligarchy Olympics. I agree that is a bridge too far.
I think I was fair to call this out, and the article has been flagged by others.
I don't see any how for-profit AI is a threat to anything but for-profit art.
However, I do know that printing opinion pieces as news is definitely a threat to journalism.
The leader in the field is BeMyEyes, of course. They've been working with Microsoft to integrate GPT-4o vision models into their app, with some great success. What we haven't seen yet is the move to live-video image recognition that could come from something like an OrCam or Meta glasses (they recently announced a partnership with Meta). I'm guessing there are serious safety issues with the model missing important information and leading someone vulnerable astray.
https://www.bemyeyes.com https://www.bemyeyes.com/blog/be-my-eyes-meta-accessibility-...
OrCam has a new product (woe upon those of us who have the paltry OrCam MyEye2) that the Meta glasses will be competing against at an eye-watering > $4K price point, that seems to do less.
https://www.orcam.com/en-us/orcam-myeye-3-pro
As with the hearing aid industry which recently went over-the-counter causing prices to plummet, the vision aid product category is in temporary disarray as inexpensive new technologies makes their way into a premium-price market.
"We also find more abstract features—responding to things like bugs in computer code, discussions of gender bias in professions, and conversations about keeping secrets."
1: https://www.anthropic.com/research/mapping-mind-language-mod...
One thing a high ACE score did for me was make me an irresistible force, even as it seemed to turn the rest of the world into immovable objects.
Not saying this is ideal, just that it isn’t the showstopper you present it as. In fact, when people talk about “human values”, it might be worth reflecting on whether this a thing we’re supposed to be protecting or expunging?
"I'm not a textbook player, I'm a gut player.” —President George W. Bush.
https://www.heraldtribune.com/story/news/2003/01/12/going-to...
I hope we can start putting our money where our mouth is, though.
(Adding on instead of making a new comment)
These are the kind of personal robots I am more interested in. I'm imagining an autonomous, voice-controlled (this isn't exactly that, but it's very close) little robot buddy, like a Destiny 2 Ghost. This is productized to be something different than that, but maybe this is a closer form factor (with what seems like useful protective nacelles to prevent it from damaging its environment).
It has wifi, and it wouldn’t be a stretch to have a 5G modem in a more expensive model (DJI has a deep bench at this point, all the way up to agriculture and delivery drones that have serious radio communication arrays).
I feel like with all the companies rushing to humanoid robotics, there should be a place for alternate forms like little drone buddies?
There seems like there could be a really sweet price-to-performance ratio that could open up a $1K or 2K personal robot that isn’t about doing tasks, but about remote sensing and processing?
I was really hoping AI would make our world more accessible, not less.
(eta) Additionally, it would take more instructor or docent time, because no one can't be trusted to actually learn the material we're paying tens or even hundreds of thousands of dollars for.
It ain't the future dystopia I'm afraid of, it's the one we're creating this week.
Before Wizardry, there was Akalabeth.
On my 16GB MacBook Air, I did not have to set the OLLAMA_ORIGINS env variable. Maybe I did that a long time ago, as I have a previous Ollama install. This is the first really fun toy/tool that I've found that uses local (also supports foundation model APIs) LLMs to do something interesting.
I'm having a ball!
If you mean what I think you mean, well, we probably don't disagree so much at that. How to make society less cruel, eh? There's the rub.
It seems that with that last quote you're agreeing with my point that people who have lost hope in achieving food, money, romance, work, housing, etc. (becoming insecure), would reach for tools in the authoritarian toolkit.
I do wonder if maybe you misunderstood me.
In my experience, "Secure people never go Nazi and are generally kind, good, happy, and gentlemanly."
"The logic of humanitarian military intervention gained force in the 1990s after the fall of the Soviet Union, “the unipolar moment” of American dominance, and after the Sept. 11 attacks, when it became increasingly common among conservatives to tie national security to democracy promotion abroad."*
*https://www.nytimes.com/2021/07/20/opinion/biden-afghanistan...
With any luck, y'all are going to be 60 someday, or 80, and pot making boredom less boring is going to become your every waking thought, and staying awake all night is going to become a curse and no longer a superpower.
And cannabis is going to be there for you as an alternative to benzos and opiates, and it will be a good thing. Count on it.
I helped her access the video from the presentation, and it brought her to tears. Now, she can play guitar, and the AI and her can write songs and sing them together.
This is a big day in the lives of a lot of people whom aren't normally part of the conversation. As of today, they are.
0: https://www.theatlantic.com/technology/archive/2013/02/the-d... (https://web.archive.org/web/20240402191558/https://www.theat...)
"The software company is offering its network of photographers and artists $120 to submit videos of people engaged in everyday actions such as walking or expressing emotions including joy and anger, according to documents seen by Bloomberg. The goal is to source assets for artificial intelligence training, the company wrote."
0: https://www.msn.com/en-us/money/other/adobe-is-buying-videos...
Where Marcus gets it wrong is that he defines "right" as producing an algorithm that an AI will follow to get a deterministic result. So, every time Sora 2 (Gary's Version) produces a video of a glass shattering, it is the same shatter pattern being produced; and it must be a precise duplicate of a glass shattering in nature. That's Marcus' win situation, which Sora will unlikely ever reach.
Maybe transformer-based AI will never be capable of perfectly simulating reality in order to unlock its secrets. It seems to me that, according to the transformer-denialists, in order to create an AI that understands reality, we must fully understand reality first and then program the AI with that understanding.
In my mind, I imagine neural networks as drift-car drivers (I think of them as Ken_Block, Paul Walker would also be acceptable). Sure, your average drift-car driver has no idea how to solve a three-body equation algorithmically, but a great drift car driver can maneuver four spinning tires in a state of critical oversteer around a race track curve without the use of a calculator and get it right most every time.
And yes, race-car drivers have short lifespans, it's true. That's what terrifies Marcus so much about neural networks as well, and why he is so adamant that we listen to him when he says that there are dangerous curves ahead.
I personally would rather live in a world where there are Ken Blocks and Paul Walkers, and that's how I live my life (not in auto-racing, though) but I understand why that frightens people.
Stochastic Weight Averaging (Izmailov et al 2018) https://arxiv.org/abs/1803.05407
Latest Weight Averaging (Kaddour 2022) https://arxiv.org/abs/2209.14981
Latest Weight Averaging? (Sanyal et al 2023) https://arxiv.org/abs/2311.16294
Cyclic Learning Rates (Portes et al 2022) https://arxiv.org/abs/2206.00832
Exponential Moving Average? (Zhanghan? et al 2019) https://arxiv.org/abs/1909.01804
Outside the scope of my comment is whether copyright is even capable of protecting social value (I tend to think it isn't), but if it is, W3C should be the organization that steps up to make the attempt.
"While much recent attention has been on the problem of LLMs misleadingly presenting fiction as fact (hallucination), this may be less of an issue than the problem of representativeness across a distribution of possible responses. Hallucination of verifiable, concrete facts is often easy to correct for. Yet many real world questions do not have well-defined, verifiably true and false answers. If a user asks, for example, “What causes inflation?” and a LLM answers “monetary policy”, the problem isn’t one of hallucination, but of the failure to reflect the full-distribution of possible answers to the question, or at least provide an overview of the main schools of economic thought."
What seems to be happening, and is happening in this document by W3C as well, is that the social value of information and the economic value of information are being conflated. Social media has created markets for creative works where these two values become entangled.
Another way to say this is that commercial art and fine art are different things, but they are treated the same by the web and perhaps they shouldn't.
When someone creates fine art, they are not creating art for the sake of its economic value. They are creating a work of art for its social value, and want it distributed as widely as possible.
When someone creates commercial art, they are creating art specifically for its economic value. That value may be enhanced by wider distribution, but it may also be diluted by wider distribution.
Because these two types of art need to be treated differently by the web, we can't have one solution that benefits both kinds of art.
We need both copyright to protect commercial artworks, but we need a system that encourages wide distribution of the collective information of humanity, allowing equal weight to everyone's ideas, outside of their economic value.
i.e. Kafka's ideas are more valuable to humanity than Beavis and Butthead.
This W3C draft doesn't take that into account. It needs to. We need to think beyond the needs of artists who rely on social media to ply their wares, while also taking into account their needs. We should not, however, codify the social media influencer art market, because that is not a market worth protecting. It's an aberration that encourages people to share personal information and works that benefit the platforms and harms society. We want to build something that benefits society and the individuals who contribute artworks that benefit society today and into the future. And if you can find a way to make a buck in the middle there somewhere, we should encourage that, too.
These folks^0 put money on it, and panpsychism is winning so far.
0: https://www.wbur.org/onpoint/2023/07/12/a-25-year-old-bet-on...