colinmorelli

This scenario seems very different because nobody gave the person any material non-public information at all, they simply deduced it from their experience participating in the trial. This feels similar to the question of "can a passenger on the Boeing jet with the door plug that blew out trade on that information" to which the answer appears to be yes.

Misappropriation theory is the following: > The misappropriation theory of insider trading is a form of insider trading where an individual trades stock in a corporation, with whom they are unaffiliated, on the basis of material non-public information they obtained through a breach of a fiduciary duty owed to the source of the information

The important part, which you're right was unclear in my comment is that the recipient of the information must have a fiduciary relationship with the source of the information, even if they do not have one with the company in question at all. That's the distinction.

This comment is not really correct

1. The misappropriation theory of insider trading covers anyone who trades on material non public information sourced through a trusted relationship regardless of any fiduciary duty to the company. For example, if I tell my personal attorney a non public fact about the company I work at, and they trade on that information, they absolutely can be found guilty of insider trading despite having no relationship to the company at hand.

2. Congress is explicitly covered by insider trading law, which was affirmed in the STOCK Act of 2012. The fact that they’re rarely indicted has more to do with the legal and political challenges associated with doing so, not the legality of the act.

OpenAI has over 5,000 employees. In addition to their headcount, they are now the highest valued private company in the world.

They're going to have some leverage.

Fair, but I'd like to clarify that in my comment I had asked specifically for any sources indicating that the PFA limits were put in place by the prior administration, since you had made the claim:

Trump's EPA created these PFAS rules

Your response was what I perceived to be a snarky comment that if only I had bothered to look, I'd have found the evidence, followed by a link that didn't say what was suggested.

Look how many people on this thread actually believe that the Trump administration is literally trying to poison them. That's not crazy to you?

The claims made all over the place are insane to me. Yes, I doubt the Trump administration is actually trying to kill me. The world is not as polarizing and extreme as people on the internet want to make it sound like it is. Most people are far more docile in the real world, but the collective hive of the internet exacerbates tension. I have no clue what side of the political aisle you're on, but my guess is we probably agree about more things than we disagree about, if we could detach bullshit labels from it all.

But FWIW, the allegation that I wasn't bothering to learn or see if I'm wrong just raised tension further. I was genuinely trying to determine if the claim was true, the evidence I had found suggested it wasn't, and it seems like it in fact wasn't quite true, but perhaps that wasn't the point you were trying to make anyway.

All fine. My hope is that we can all turn down the tension and hostility a level or two. Might be the only hope we have.

If you'd truly like to learn if you're wrong, it's recommended to seek information that disproves your hypothesis rather than proves it. Both this and the previous article I shared were very easy to find and within the first 2 or 3 results.

Firstly, this is a completely unnecessary comment. My searches were specifically regarding finding the enactment of specific PFA limits. I will acknowledge to not spending that much time looking at it, as you claimed to already have a source and I was curious to see what it was.

But to the point, this document does not outline or set limits on PFAS in drinking water. It's an action plan for measuring and creating limits, but does not itself enforce anything. In fact, every subsequent search I've done has shown that the 2024 Final Rule was the first point at which any limits were put into action.

Quoting directly, the document states that one of the steps being taken is:

Initiating steps to evaluate the need for a maximum contaminant level (MCL) for perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS);

In other words, it outlines a plan for the research that is used to 1) determine if MCL should be set, and 2) what, if any, it should be set to. Notably, it does it not itself set that limit or come to a conclusion about what it should be.

Further, this research appears to be a continuation of research released in 2016 [1], which was the first time that a guideline (but not a mandate) was set. This would, of course, be prior to Trump's first administration. This is suggested in the document itself, where it outlines that this document is part of a series of actions beginning in 2015/2016, as well as callouts to specific research in the 2016 article linked below.

So the facts seem to show that: 1) The first guideline was set in 2016. It was not a law at this time. 2) Research continued to identify next steps for setting a standard, which were codified and shared in the 2019 article you linked 3) The 2024 Final Rule put a MCL into action for PFAS.

Take from that chain of events what you will, but the initial accusations of "political bias" seem unfounded here.

[1] https://www.epa.gov/sites/default/files/2016-05/documents/pf...

It seems like the PFAS rules were set in prior administrations [1]. In fact, even in the article you've linked above, the text states:

retaining its maximum contaminant levels for PFOA and PFOS but pulling back on its use of a hazard index and regulatory determinations for additional PFAS

Key word being "retaining," indicating the maximum contaminant levels were already in place prior to the change mentioned here. Putting aside allegations of "political bias," can you point to a source which clearly indicates the PFA limits were put in place by the current administration? Would like to learn if I'm wrong.

[1] https://www.epa.gov/newsreleases/biden-harris-administration...

in this specific case

You could have quoted the beginning of the sentence, where the point was about this specific case, and how in this particular case, a gun clearly allowed an assassination that would have been challenging to pull off with a knife.

That is not a way as saying killing someone with a knife is impossible. It's a way of saying that guns allow you to kill people in ways and distances that knives do not.

Yeah I'm not suggesting the same process could apply in the US, I'm just trying to aggressively refute the point that guns are not the problem (or, at least, a major component of it). We need to be creative about solutions, but people have to want to find a solution to be creative about them, and right now many do not.

This narrative isn't helpful. Even in this specific case, it's extremely unlikely anyone would have been able to get close enough to him with a knife to kill him without someone noticing.

Guns allow you to kill 1) multiple people, 2) from a distance, and 3) with nobody aware of the imminent threat.

Of course other weapons can also be used to harm people. Of course no solution is perfect. But it's absolutely incorrect to say "the problem isn't so much the tools." The tools undeniably and irrefutably play a role in every study that has ever been conducted on this topic.

See here for the impact of Australia's gun buyback program, which saw zero mass shootings in a decade after their removal, after 13 mass shootings in the 18 years prior the removal, as well as an accelerated decline in firearm deaths and suicides: https://injuryprevention.bmj.com/content/12/6/365

I'll take a shot at rationale for this perspective, which is similar to a peer comment:

The tech is undoubtedly impressive, and I'm sure has a ton of headroom to grow (although I have no direct knowledge of this, but I'd take you at your word, because I'm sure it's true).

But at least my perception of the idea that this is a "bubble" presently is rooted in the businesses that are created using the technology. Tons of money spent to power AI agents to conduct tasks that would be 99% less expensive to conduct via a simple API call, or because the actual unstructured work is 2 or 3 levels higher in the value chain, and given enough time, there will be new vertically integrated companies that use AI to solve the problem at the root and eliminate the need for entire categories of companies at the level below.

In other words: the root of the bubble (to me) is not that the value will never be realized, but that many (if not most) of this crop of companies, given the amount of time the workflows and technology have had to take hold in organizations, will almost certainly not be able to survive long enough to be the ones to realize it.

This also seems to be why folks draw comparison to the dot com bubble, because it was quite similar. The tech was undoubtedly world changing. But the world needed time to adapt, and most of those companies no longer exist, even though many of the problems were solved a decade later by a new startup who achieved incredible scale.

To be fair, if those people are right, then NVIDIA's stock price (and revenue) is part of that bubble, so its not really evidence that this isn't a bubble.

Time will tell if they're right or not. But it wouldn't be the first time it has happened.

Both things can be true. The tech can be transformative, and the current valuations and burn rate can be wholly unsustainable in the short term. This is exactly what happened in the dotcom era.

Whether or not this one is the same is impossible to know until after it happens. But there are credible arguments to both sides.

Even if we accept as a premise that these models are doing "smart retrieval" and not "reasoning" (neither of which are being defined here, nor do I think we can tell from this tweet even if they were), it doesn't really change the impact.

There are many industries for which the vast majority of work done is closer to what I think you mean by "smart retrieval" than what I think you mean by "reasoning." Adult primary care and pediatrics, finance, law, veterinary medicine, software engineering, etc. At least half, if not upwards of 80% of the work in each of these fields is effectively pattern matching to a known set of protocols. They absolutely deal in novel problems as well, but it's not the majority of their work.

Philosophically it might be interesting to ask what "reasoning" means, and how we can assess if the LLMs are doing it. But, practically, the impacts to society will be felt even if all they are doing is retrieval.

1,155 is the number of employees in the research office. The EPA overall had 16,155 employees in January 2025 [1].

[1] https://www.epa.gov/newsreleases/epa-announces-reduction-for...

It is absolutely possible for the unit economics of a product to be profitable and for the parent company to be losing money. In fact, it's extremely common when the company is bullish on their own future and thus they invest heavily in marketing and R&D to continue their growth. This is what I understood GP to mean.

Whether it's true for any of the mainstream LLM companies or not is anyone's guess, since their financials are either private or don't separate out LLM inference as a line item.

This, similar to your other comment, is unrelated to my comment.

This is about determining if AI can be a equivalent or better (defined as: achieving equal or better clinical outcomes) therapist than a human. That is a question that can be studied and answered.

Whether artificial intelligence accurately models human intelligence, or whether an airplane is "smarter" than a bird, are entirely separate questions that can perhaps serve to explain _why/how_ the AI can (or can't) achieve better results than the thing we're comparing against, but not whether it does or does not. Those questions are perhaps unanswerable based on today's knowledge. But they're not prerequisites.

I'm not exactly sure how this relates to my comment above. An analysis of an airline crash and a study are not the same thing.

As it relates to study design, controlling for set and setting are part of the methodology. For example, most drug studies are double-blinded so that neither patients nor clinicians are aware of whether the patient is getting the drug or not, to reduce or eliminate any placebo effect (i.e. to control for the "set"/mental state of those involved in the study).

There are certainly some cases in which it's effectively impossible to control for these factors (i.e. psychedelics). That's not what's really being discussed here, though.

An airline crash is an n of 1 incident, and not the same as a designed study.

This is an interesting take. By this perspective, it's essentially impossible to ever gauge the efficacy of AI in doing anything, because the people who will know how to measure the quality of that thing are also the people who will be displaced by showing the AI can do that thing. In fact, you could probably argue that every study ever is worthless, because studies are generally performed by people who know the subject matter and it's basically impossible to be unbiased on a topic if you're also highly knowledgable about said topic.

In reality, what matters is the methodology of the study. If the study's methodology is sound, and its results can be reproduced by others, then it is generally considered to be a good study. That's the whole reason we publish methodologies and results: so others can critique and verify. If you think this study is bad, explain why. The whole document is there for you to review.

You conveniently left out the first part of the sentence you quoted:

Currently available software may very well make human drivers both more comfortable and safe...

Which is objectively not what Waymo does, and whether intentional or not, invalidates the progress that has been made.

Also, immediately preceding that:

Driverless vehicles in closed systems have been in use for a long time.

Which is also not what current frontier self driving technology is.

Where did anyone make this argument?

The title of the article is quite literally "Is Winter Coming?"

Sorry, I should have clarified, but no this is not ChatGPT's self assessment.

I am suggesting that today's best in class models (Gemini 2.5 Pro and o3, for example), when given the same context that a physician has access to (labs, prior notes, medication history, diagnosis history, etc), and given an appropriate eval loop, can achieve similar diagnostic accuracy.

I am not suggesting that patients turn to ChatGPT for medical diagnosis, or that these tools are made available to patients to self diagnose, or that physicians can or should be replaced by an LLM.

But there absolutely is a role for an LLM to play in diagnostic workflows to support physicians and care teams.

For what it's worth this statement is actually not entirely correct anymore. Top-end models today are on par with diagnostic capabilities of physicians on average (across many specialties), and, in some cases, can outperform them when RAG'd in with vetted clinical guidelines (like NIH data, UpToDate, etc)

However, they do have particular types of failure modes that they're more prone to, and this is one of them. So they're imperfect.

The point is not that those things were meant when we said self driving cars. It's that, at every step along the way, there were a group of people who doubted that cars could do that thing, and then they did that thing. And then the thing we said they can't do changed to something else.

Today, you absolutely can "get in a car, tell it where you want to go, and it goes there while you read a book" - it's literally what Waymo is and has been doing. And now we're saying it can't do it in Mumbai, so it's still not self-driving.

At some point, the distinction seems pointless. We are undeniably continuing to make progress on the road to autonomous driving, and it does work in certain scenarios today. To suggest things are slowing down because we haven't met the most reason interpretation of the words is neither helpful nor correct.

Related similar thing when I sent my dog's recent bloodwork to an LLM, including dates, tests, and values. The model suggested that an advancement in her kidney values (all still within normal range) were likely evidence of chronic kidney disease in its early stage. Naturally this caused some concern for my wife.

But, I work in healthcare and have enough knowledge of health to know that CKD almost certainly could not advance fast enough to be the cause of the kidney value changes in the labs that were only 6 weeks apart. I asked the LLM if that's the best explanation for these values given they're only 6 weeks apart, and it adjusted its answer to say CKD is likely not the explanation as progression would happen typically over 6+ months to a year at this stage, and more likely explanations were nephrotoxins (recent NSAID use), temporary dehydration, or recent infection.

We then spoke to our vet who confirmed that CKD would be unlikely to explain a shift in values like this between two tests that were just 6 weeks apart.

That would almost certainly throw off someone with less knowledge about this, however. If the tests were 4-6 months apart, CKD could explain the change. It's not an implausible explanation, but it skipped over a critical piece of information (the time between tests) before originally coming to that answer.

I'm not sure I get your analogy here. If you're suggesting that it's not "moving the goalposts" because it's pointing out that driving in Rome or Mumbai is different than driving in North America, then that is exactly what is meant by moving the goalposts.

10 years ago the claim was that "cars can't drive autonomously," Waymo quietly chips away to the point that they absolutely can drive autonomously, even in an unpredictable environment (with evidently drastically lower-than-human accident rates, for example), and the reaction of those original people is to say "yeah but it can't drive in [even more complex place]"

Sure, that's not exactly surprising. We generally don't design technology to do the most complex version of the task it's supposed to do first. We generally start with a simpler scenario it can accomplish and progressively enhance it as we learn more. Cars have been doing that for decades.

So perhaps the tech doesn't work in Mumbai or Rome yet. Maybe we'll advance the tech to do that thing, or maybe we'll come up with a different solution to autonomous driving in these places if we find out it'll be more expensive to advance this technology than it will be to do something else instead. But either way, it's already doing the thing that many, many people claimed it can't do, and those people are now claiming there's something else it can't do. That is the very definition of moving the goalposts.

This feels a lot like "moving the goalposts." First, it was complete science fiction to have technology in the car. Then, it was in the car, but it could only do navigation and music, it can't operate the car the way humans can. Then, it can prevent you from weaving out of your lane, and it can stop the car if you're about to crash into something, but it can't help you with your commute. Then, it can speed up, slow down, and steer on the highway, but it can't take you door to door. Now, it can take you door to door, but only in certain environments, it can't do it everywhere.

All of the above happened over the last ~20 year or so. The progression clearly seems to point to this being more than hype, even if it takes us longer to realize than originally anticipated.

I find this way of looking at LLMs to be odd. Surely we all are aware that AI has always been probabilistic in nature. Very few people seem to go around talking about how their binary classifier is always hallucinating, but just sometimes happens to be right.

Just like every other form of ML we've come up with, LLMs are imperfect. They get things wrong. This is more of an indictment of yeeting a pure AI chat interface in front of a consumer than it is an indictment of the underlying technology itself. LLMs are incredibly good at doing some things. They are less good at other things.

There are ways to use them effectively, and there are bad ways to use them. Just like every other tool.

"Hallucination" implies that the LLM holds some relationship to truth. Output from an LLM is not a hallucination, it's bullshit[0].

I understand your perspective, but the intention was to use a term we've all heard to reflect the thing we're all thinking about. Whether or not this is the right term to use for scenarios where the LLM emits incorrect information is not relevant to this post in particular.

No we don't. It's really complicated. That's why diets are popular and real dietitians are expensive.

No, this is not why real dietitians are expensive. Real dietitians are expensive because they go through extensive training on a topic and are a licensed (and thus supply constrained) group. That doesn't mean they're operating without a grounding fact base.

Dietitians are not making up nutritional evidence and guidance as they go. They're operating on studies that have been done over decades of time and millions of people to understand in general what foods are linked to what outcomes. Yes, the field evolves. Yes, it requires changes over time. But to suggest we "don't know" is inconsistent with the fact that we're able to teach dietitians how to construct diets in the first place.

There are absolutely cases in which the confounding factors for a patient are unique enough such that novel human thought will be required to construct a reasonable diet plan or treatment pathway for someone. That will continue to be true in law, health, finances, etc. But there are also many, many cases where that is absolutely not the case, the presentation of the case is quite simple, and the next step actions are highly procedural.

This is not the same as saying dietitians are useless, or physicians are useless, or attorneys are useless. It is to say that, due to the supply constraints of these professions, there are always going to be fundamental limits to the amount they can produce. But there is a credible argument to be made that if we can bolster their ability to deliver the common scenarios much more effectively, we might be able to unlock some of the capacity to reach more people.

As for correctness, they mentioned the LLM citing links that the person can verify. So there is some protection at that level.

But, also, the threshold of things we manage ourselves versus when we look to others is constantly moving as technology advances and things change. We're always making risk tradeoff decisions measuring the probability we get sued or some harm comes to us versus trusting that we can handle some tasks ourselves. For example, most people do not have attorneys review their lease agreements or job offers, unless they have a specific circumstance that warrants they do so.

The line will move, as technology gives people the tools to become better at handling the more mundane things themselves.

This is the real value of AI that, I think, we're just starting to get into. It's less about automating workflows that are inherently unstructured (I think that we're likely to continue wanting humans for this for some time).

It's more about automating workflows that are already procedural and/or protocolized, but where information gathering is messy and unstructured (I.e. some facets of law, health, finance, etc).

Using your dietician example: we often know quite well what types of foods to eat or avoid based on your nutritional needs, your medical history, your preferences, etc. But gathering all of that information requires a mix of collecting medical records, talking to the patient, etc. Once that information is available, we can execute a fairly procedural plan to put together a diet that will likely work for you.

These are cases that I believe LLMs are actually very well suited, if the solution can be designed in such a way as to limit hallucinations.

FYI: The actual study may not quite say what this article is suggesting. Unless I'm missing something, the study seems to focus on employee use of chat-based assistants, not on company-wide use of AI workflow solutions. The answers come from interviewing the employees themselves. There is an analysis of impacts on the labor market, but that is likely flawed if the companies are segmented based on employee use of chat assistants versus company-wide deployment of AI technology.

In other words, this more likely answers the question "If customer support agents all use ChatGPT or some in-house equivalent, does the company need fewer customer support agents?" than it answers the question "If we deploy an AI agent for customers to interact with, can it reduce the volume of inquiries that make it to our customer service team and, thus, require fewer agents?"