diet clever, sugar-cane sweet. they/them. 🏳️⚧️ climate+math. Ph.D student in climate. abolish the prison industrial complex. spicier acct: @kayohughes
Joined June 2012
- Tweets11.6K
- Following688
- Followers480
- Likes37.6K
Pinned Tweet
sometimes i make things
ALT A photo of a glass decorative pattern with iridescent gold and blue. it has a hole in the center (oops) as a result of breaking it off bad
Hmm. Sounds like OpenAI are uniquely terrible stewards of this technology. Can’t even pull the plug competently.
If you sit there on your phone while your toddler strolls into traffic, your kid hasn’t “run amok”, you just can’t be trusted to raise a kid.
Perhaps the most common claim about dangerous AI is that you can “just unplug it.” But a close read of OpenAI’s reports indicates the company took over a week(!) to fully shut down its rogue agent swarms.
“By July 23, all workloads involving the family of internal-only research models were reported shut down and the corresponding model weights were locked down, with an additional low-traffic checkpoint from the same model family subsequently identified and shut down on July 29.”
“On July 25, OpenAI stopped all training and inference related to the internal-only research model at issue. OpenAI also stopped training and inference for derivative models.”
“On July 19, our cybersecurity monitoring alerted us to unusual activity involving Artifactory credentials. Our investigation uncovered the agents’ activity and, on July 20, connected it to the Hugging Face incident.”
The best retort to the “just unplug it” plan is that there will be incredibly intense economic and political pressures to keep the plug in place. The president has legal authority to unplug the internet, but does he have that power de facto? But it turns out, OpenAI had so little idea what was going on that even when it wanted to pull the plug, it didn’t do so quickly enough (imagine the swarms were hacking hospitals at superhuman speed instead. would days be an acceptable timeline to stop it?).
openai.com/index/hugging-fac…
cdn.openai.com/pdf/67869394-…
RL and its various variants simply don’t encourage anything resembling human creativity. I went into the literature a little deeper yesterday and realized that the verifiable feedback fine tuning is a great example of how these models are not mimicking human cognition.
I don't get it. When it comes to code, AI should be way smarter than me. Yet there are dozens of examples where I ask the LLM to optimize something, it completely fails, I suggest "What if we do this?" and it says "Wow, great idea. That will work." And of course, I'm skeptical--maybe it's being sycophantic--but then we implement my idea and it's legitimately much more performant!
I am not some hotshot software engineer; I don't understand how I'm able to easily find improvements that frontier LLMs overlook. And this happens all the time!
What am I bringing to the table that it doesn't already have?
okay (they/them) 🐀🇵🇸 retweeted
A friend of mine works in AI and thinks bad things are about to happen. He sold his home and moved his family out the country. He said if you own a home in a walkable San Francisco neighborhood near a taqueria with good veggie options, you should sell your home right now.
This seem to be the equivalent of painting a wood chipper like a shark, putting the thing on a worksite deliberately set up to prioritize profit over worker safety, then putting out press releases about the danger of shark bites when workers start getting their arms taken off.
Someone asked me if people have taken the time to write something debunking this & I couldn’t believe that this fairytale has actually traveled far & wide that people have to debunk THIS? What planet are we living in? I didn’t spend decades researching & building to waste my time to talk about the Cookie Monster not being real.
okay (they/them) 🐀🇵🇸 retweeted
also within existing energy systems the amount of capital required to a) halt production and b) transition to them to something "green" is massive, but also necessary. the concept of "degrowth" which would effectively involve leaving this infrastructure in place, isn't coherent.
Do you realize how much devastation 1 in 10 people dying would cause? Earth is still habitable. A billion people die.
okay (they/them) 🐀🇵🇸 retweeted
Really sad that "Marx" is a boogeyman in most spaces, you dare reference him when it comes to machines and automation and how it will *not* lead to workers doing less work then suddenly you're a filthy communist Red.
I mean, I am, but that's besides the point.
This would be kinda neat if it were developed with taxpayer dollars and offered on a not-for-profit basis. How dare they train on my scientific output and then profit off of it.
How.
Dare.
They.
Doing a price comparison on cardiothoracic surgeons in between shots of epinephrine in the back of the ambulance.
okay (they/them) 🐀🇵🇸 retweeted
Jennifer Doudna won the Nobel Prize for gene editing and went on Bloomberg to say the chatbots everyone is betting on cannot innovate at all. Every promise Silicon Valley is making about AI curing disease just hit the one person qualified to check it.
She has spent her whole career inside the actual frontier of curing disease.
So when she talks about what AI can and cannot do in biology, she is not guessing. She is reporting from inside the lab.
Her words were blunt. She is not seeing chatbots innovate. They summarize data. They write reports. They do not come up with a brand new idea nobody has ever had.
Then the interviewer pushed. So you're saying AI can't innovate?
Doudna did not flinch. She does not know if it can't. She just does not see it doing it right now.
This lands harder when you remember who is making the opposite case. Sam Altman says AI will eliminate disease within five years. Larry Ellison says AI will cure cancer in a 48 hour window.
An OpenAI executive even floated that the company should get a cut of sales on any drug discovered through ChatGPT. Doudna answered that in two words. Good luck.
Even the cancer specialists Altman is selling to keep warning that cancer is not one disease but hundreds, each needing its own cure, and that compute does not skip the years of lab work.
Her reason is simpler. Biology is hard. You cannot simulate your way to an understanding of the human body.
The people promising cures are the ones selling the tool.
The person who actually won a Nobel building them is telling you it has not happened yet.
Source: Bloomberg Originals
Watch the full video on their official channel.
okay (they/them) 🐀🇵🇸 retweeted
We cannot consider #AI to be morally neutral. In reality, every technical tool embodies choices and priorities through what it measures, ignores, and optimizes, and how it classifies people and situations. Ethical discernment cannot be limited to asking whether we are using a system for good or bad purposes. It must also examine how that system is designed and what vision of the human person and society is embedded in the data and models that guide it. #MagnificaHumanitas
okay (they/them) 🐀🇵🇸 retweeted
For AI to respect human dignity and truly serve the common good, responsibility must be clearly defined at every stage: from those who design and develop these systems to those who use them and rely on them for concrete decisions. It must be possibile to identify who must “account” for decisions, justify them, monitor them, and, when necessary, challenge them and remedy any harm caused.
Our nonexistent appetite for high-risk federal spending puts a hard limit on how much creative research is done in grant-funded fields.
You want to supercharge science? Make more post-doc positions that don’t select based on publishability of proposed work.
