>Let’s say you need to get your oven fixed—Alexa+ will be able to navigate the web, use Thumbtack to discover the relevant service provider, authenticate, arrange the repair, and come back to tell you it’s done—there’s no need to supervise or intervene.
This is a disaster waiting to happen. I don't trust an LLM to choose between two brands of dish soap for me let alone pick a contractor, schedule a repair, and make a payment. Even if there was a demo showing this working in a sterile environment, reality is so complex that something is certain to go wrong. Even the "simple" task of summarizing news had so many catastrophic failures that Apple had to pull it from the market.
Amazon is making bold claims about the capabilities of their voice assistant to sell their subscription service so that they can make the Alexa division profitable, but if any of their claims were real, they would be demoing rather than writing science fiction in a press release.
It’s worse than that because SEO excreta is still shown to the user, who at least has a chance to question it. With this style of assistant, the user doesn’t even see what the bot saw and that opens up new areas for abuse which is harder to detect.
This reminds me of Facebook's "M" assistant from 2017.
> Today, a few hundred Bay Area Facebook users will open their Messenger apps to discover M, a new virtual assistant. Facebook will prompt them to test it with examples of what M can do: Make restaurant reservations. Find a birthday gift for your spouse. Suggest---and then book---weekend getaways.
Just like nuclear fusion is always 30 years away, you can find a handyman and schedule an appointment with a voice assistant / chatbot in 6 months, and everyone will use self driving cars in 5 years.
I was a little kid when a local university came out with news that they had a big breakthrough in Cold Fusion. I remember my mom telling me it was going to change everything and we would travel around the world on invisible rails.
Many years later the worlds richest man is digging short tunnels to drive electric vehicles in.
My house, not too far from Three Mile Island, was built between its commissioning and the incident with reactor 2. As a result, its builders installed the most insane heating system known to man: resistive heating in the ceiling. Crank it all you want, your feet will still be cold. It’s a system that could only possibly be acceptable if electricity doesn’t cost anything.
They weren't unreasonable - they just didn't predict the whole world going collectively insane after Chernobyl, and choosing to pay for being slowly suffocated and cooked to death instead.
It also seems like it's ripe for just being an outright lie. People will pay Amazon to be ranked as the preferred service provider. You won't get the best service provider, you'll get the one that paid Amazon the most money.
They already failed at this. Adding something to a shopping cart is just one of the steps the mentioned example states. And consumers have already indicated that they don't trust it, because they prefer to know just what they're buying. So this project, by doubling down on that, is DOA
Yeah there's roughly zero chance that works reliably. It's so prone to bad failure cases I'm skeptical that they'll even ship something that tries to do that automatically.
On the plus side if they do ship this we should get all sorts of amusing stories out of it. I'm picturing someone saying offhand "I'm so hungry I could eat a horse!"
I'd expect this to be like when you call the bank and go through the phone tree, enter your account, birth date, SSN, zip code, etc. Finally get connected to the rep and the first questions are "What's your name and birth date?"
> Even the "simple" task of summarizing news had so many catastrophic failures that Apple had to pull it from the market.
Unsurprisingly this human-made summary of Apple’s missteps is grossly incorrect as well. There were no catastrophic failures and the feature being poorly implemented does not warrant the hyperbole.
i can picture your ancestor saying "what? i will go to one mega store and buy something made in asia instead of getting a suit made to measure by Giovanni across the street?"
or your mom complaining about "buying shoes online without trying then" (or maybe that's you if born before 2000)
AI slop will define your identity tomorrow, doesn't matter what you think
Let me put this way: as a non native English speaker, I am fairly confident that I'll do a better job at "summarizing" headlines than Apple Intelligence. Take that however you like.
And there are other people living in the same zip code as you. A single family may even have more than one child living under the same roof. Naturally the question will arise anyway as to why there are differences in learning speed or fluency.
Seems high to me, but I've never taken the drug. I assume most people take it just a few times a week for it's intended use? Is anyone aware of research that shows positive benefits to health at recreational doses?
That’s a standard dose. 3X per day is because it has a short half life.
> Is anyone aware of research that shows positive benefits to health at recreational doses?
Recreational doses? What does that mean? Excessive vasodilation isn’t going to be fun. You actually need some vasoconstriction to balance your blood pressure. Too much and you’re just going to get headaches and lightheaded. Excessive vasodilation would be expected to worsen cognition.
Giving the parent the benefit of the doubt, they probably were confused by the idea of people taking these pills to get high. That's generally what is meant by the term "recreational use" when it comes to drugs.
"no major AI technology breakthroughs in decades.everything we are seeing is larger compute scaling." This is false. Everything from the transformer to advancements in state space models have been foundational breakthroughs
People who say things like "major breakthroughs" often imagine a cliff/steep rise. The reality is that most "breakthroughs" are small, incremental, almost invisible steps in all aspects of the field, and potentially even in fields that are only tangentially related.
Then, one day, they hear about it in the news, because there's now some hype, or some event that makes it newsworthy. This makes it feel like the breakthrough was instantaneous or steep, but in fact has been in the making for decades or even more.
I have no doubt AGI is coming, but it will be gradual and slow. It will be the accumulation of more advances in everything, including hardware, as well as software. It might even include economic changes.
From nothing to ChatGPT opened to the public and billions students' lifes rely on that in the next month, isn't this the biggest step in the tech history?
It's a major societal failure that education has been reduced to turning in coherent enough strings of characters, that's what I've learned in the past year.
You mean the use of a writing system to share knowledge amongst ourselves?
I find that absolutely wonderful and it worked decently well for me (and possibly you.) Now we have a never seen before technology and society will adapt, that's it. No failure.
> You mean the use of a writing system to share knowledge amongst ourselves?
No. And what's with the "it is so bad yet you used it"? I am very much allowed and required to denounce a system even if I cannot escape it or if I could have somehow profited from it or chosen to use it.
I very much reached the point where I am despite the educational systems I was exposed to. And a system geared towards memorization and regurgitation of data in textual format where pupils can successfully use a chatbot to avoid doing work is certainly failing its goals of educating the youth.
I would point you to the complaints of American teachers about the reading and mathematics levels of students, if only because that is widely accessible. I did not grow up in the US and the school system in my home country is leagues behind the US.
excluding... antibiotics, electricity, refrigeration, the combustion engine, the digital computer...
like even restricted to the domain of strictly computation, I'd say it barely scratches the surface... like even if we ignore computer engineering ("the transistor," "silicon microprocessors", etc.)... foundational tech like "compilers" are more significant.
even restricted to modern applications, GPS is more useful and life-changing.
so, no. It's not the biggest step in tech history.
Curious about how many kids in India and China are relying on it.
I'll accept the prospect of hundreds of millions, generously.
And "relying on it" is a strong phrase. Using it as a curiosity, sure. The ones relying on it seem to keep ending up in the news because of how, well, unreliable it is.
There were hundreds of competitive, even SOTA LLMs before ChatGPT existed. You're basically just proving the parent comment right in how small of a leap ChatGPT is from t5-flan or BERT.
I beg to differ. Transformers are purely an optimization. It’s not exactly right to call everything “compute scaling” but we are still, at the end of the day, fitting polynomials.
And frankly, that’s probably not what our brains are doing.
It is a bit weird though… I mean, you could just as well say that the last breakthrough in all of computing was the transistor (Shockley 1946) – all we’ve been doing ever since is just “scaling” and combining them in new ways.
AI has been a 'moving aim' as opposed to moving target. It's a label slapped on whatever slice of software engineering seeming most magic-like at any given decade.
Early computers in late 40s were called 'electronic brains' by the media...
It’s difficult to comprehend impact of something that’s widely adopted. Like, batch normalization alone was probably mind blowing when it came out. Yet it seems so simple and self-explanatory now.
> And frankly, that’s probably not what our brains are doing.
I think that it is! It's much more likely to me that our brains are doing something big and simple than small and complicated. That's the way that nature tends to work. Fitting low-order million-dimensional polynomials would meet that description.
Our brain is however the very definition of small and complicated. It's the most complex known organic object known to humans and for all that any given normal brain handles in a given day, it uses just 0.3 kilowatt hours (kWh) to do it. No computer we have comes close to handling what a brain handles with that power consumption.
ChatGPT by contrast, consumes roughly a gigawatt hour per day serving its users. Yes, to do this it's handling a colossal amount of queries, but that's all it's doing. Your brain handles everything in your body and consciousness, in ways we don't even fully understand, while also letting you think and communicate and reason as a conscious being with self direction.
Moreover, there is evidence that at least part of our brain's functions may be exactly as the other reply here mentions, weird, subatomic and deeply complex in ways that are difficult to get a clear grip on.
From the double slit experiment, to particle-wave duality, to the particle zoo of the 70s, to quantum chromodynamics, to asymptotic freedom, to more exotic theories like string theory, etc. tells us the complete opposite. Every major discovery in physics in the past 150 years seems to disagree. Things are extremely weird and complicated when we get extremely tiny. Why would our brains be different?
If we lived at the quantum scale, then classical physics would be the weird one. Quantum chromodynamics is only confusing for two reasons: it differs from our everyday experience so we don't have an intuition for it, and because it has a large number of mutually-interacting (but basic) components.
Richard Feynman put it very well:
"The world is strange, the whole universe is very strange, but see when you look at the details then you find out that the rules are very simple, of the game, the mechanical rules by which you can figure out exactly what's going to happen when the situation is simple. It's again this chess game; if you're in just the corner with only a few pieces involved, you can work out exactly what's going to happen. And you can always do that when there's only a few pieces. And so you know you understand it. And yet, in the real game there's so many pieces you can't figure out what's going to happen.
"There's such a lot in the world, there's so much distance between the fundamental rules and the final phenomena that it's almost unbelievable that the final variety of phenomena can come from such a steady operation of such simple rules... But it is not complicated, it's just a lot of it."
This comment is everything wrong with media literacy. It's absolutely worthwhile to cover highly public calls to violence of government officials by respected individuals with lots of power and the article makes it clear he personally did not send the letters. But denying that public calls to violence spurs actual violence is denial of basic cause and effect.
> It's absolutely worthwhile to cover highly public calls to violence of government officials by respected individuals with lots of power
I mean, sure...who are you arguing with? I didn't say nobody should cover this. I said this article is terrible.
> But denying that public calls to violence spurs actual violence is denial of basic cause and effect.
Yeah, except we have laws around this concept, and even if what you're saying were true in the US (it isn't, thankfully), it doesn't magicaly make hack journalism good.
Said differently, "incitement to violence" doesn't mean that missionlocal is high-minded and mature for spending two articles talking about the price of his liquor.
> Said differently, "incitement to violence" doesn't mean that missionlocal is high-minded and mature for spending two articles talking about the price of his liquor.
Garry Tan chose to flaunt the high-end liquor bottles and "Twitter menace" plaque ahead of his sort-of-apology, not missionlocal.
Only one of the articles (not the one linked) refers to them at all. The other one focuses more on the hate mail some idiot decided to send being a screenshot of Tan's original tweet, and both of them are pretty clear about it being a rap lyric.
Kind of hard to argue with a straight face that the real problem with the YC CEO
acting like a not-very-smart bro influencer even in his sort-of-apology is that some local rag journalist didn't spare the embarrassing detail.
Sure, Tan was probably more interested in highlighting the "twitter menace" plaque than the fairly expensive liquor and unremarkably-priced wine, but celebrities flaunting wealth with a laughing emoji as a "fuck you" to their critics is a well established trope, and I don't think high-minded, mature journalism is about taking the most sympathetic interpretation possible of bro silliness.
More necessary than the sneer quotes you applied to the "journalist", yes. If I say dumb stuff when drunk and then muff the initial apology (when presumably sober) I wouldn't expect people to sugar coat it either.
Garry clearly isn't too stupid or desperate for clicks to do any better, and so I'm afraid I'm going to have to continue to disagree that we should save our ridicule for his critics.
Surprised to see the top comments here so doom-and-gloom: "ah it was posted in 2016, I don't think anyone (but do wish) cares about limiting their JS usage anymore "
The tools we have available today to optimize JS and build for progressive enhancement are light years ahead of 2016 because the smartest minds in the industry have optimized for it. The rise of batteries-included server-side-rendered React apps like Remix and Next massively cut down on time-to-first-paint and React 18 makes it easier than ever to defer rendering of async or interactive portions of the page (e.g. the Like count, the reply form) in favor of getting static content on the screen as fast as possible.
This post has a very odd tone, as if a private company enforcing their policy to remove unverified & leaked personal correspondence of a private citizen is some unquestionable moral wrongdoing that's apparently going to blow up. Surprised to see this as at the top, on HN.
National Review is a conservative wing-nut website trying to turn this non-story into fuel for their censorship culture war.
The media commonly relies on second hand accounts of leaked documents ("persons familiar with the contents of [X] document"), including private documents.
Something like this has never been censored before.
It's not like they are taking down a post from a nobody. The NY Post is a major publication.
It's twitter saying they know better than the NY Post. That is a major step that I don't think has ever been taken before.
Not only that, but IMO context matters here. We are 20 days away from the election. People are standing in lines to vote as we speak.
If this was published at any other time (and the material for the story has apparently existed since December) the public and the professional media would have had time to scrutinize it, discuss its shortcomings, etc.
But, as Winston Churchill said "A lie gets halfway around the world before the truth has a chance to get its pants on." We can't do that and being complicit in spreading misinformation in the middle of an election is, I'm sure, still a strong memory that social media companies have from the 2016 election.
My guess as to how this will go is that it will take a week or two to authoritatively discredit the article, but by then, if left unchecked, the article will have already done its damage. The cynic in me says that the originators of the article already know that there isn't much truth here and that this is the point of releasing it while people are voting.
Love the downvotes for this. Anyone who thinks for a second that Twitter and FB have some kind of set of principles that this story and only this story have crossed is deluded.
Did twitter censor all stories regarding the claim that Trump paid $750 in taxes?
In case the point I’m making isn’t obvious: the NYT never published their source for that, and still haven’t. If twitter is removing stories for having dubious sources, then that story should not pass muster either.
If they’re removing stories for having “hacked” (or in the case of both this story about Hunter Biden, or Trumps taxes: “leaked”) sources, then discussion of BOTH of these stories should be banned.
That's a bizarre comparison, because Trump has repeatedly lies about being willing but unable to release his own copy of his tax returns. And the NY Times didn't publish personal documents, they published facts about a certain class of personal documents (tax documents of a Presidential candidate) which for decades have been considered public information by all major corners of the political arena.
Those weren't hacked or leaked. The story is much simpler, and the provenance of the files is not in question:
By falsely claiming the records were hacked, rather than the legal property of the repair shop following payment default by Hunter Biden, Twitter is itself deliberately spreading false information to justify its illegal election interference. [1]
The provenance of the files is not questioned by far-right political operatives like Sean Davis who are pushing this story in the first place, but to anybody with half a brain cell and an ounce of skepticism the whole story stinks. Just listen to this interview with the owner of the repair shop these files purportedly came from, who changes his story about a half-dozen times in the span of sixty minutes:
My pick - Daily Mail: a newspaper headquartered in London, UK - a 3rd party country, independent from the ongoing US election. Over 120 years long history of investigative journalism.
Your pick - The Daily Beast: a 12 years old opinion website in USA, directly interested in the ongoing US election.
Compare and contrast the opening - your choice focused on "character assassination" via conspiracy theory; my source focused on giving a broad background & overview as facts. The difference in quality of reporting is palpable.
> Lord Rothermere was a friend of Benito Mussolini and Adolf Hitler, and directed the Mail's editorial stance towards them in the early 1930s.[44][45] Rothermere's 1933 leader "Youth Triumphant" praised the new Nazi regime's accomplishments, and was subsequently used as propaganda by them.[46] In it, Rothermere predicted that "The minor misdeeds of individual Nazis would be submerged by the immense benefits the new regime is already bestowing upon Germany". Journalist John Simpson, in a book on journalism, suggested that Rothermere was referring to the violence against Jews and Communists rather than the detention of political prisoners.[47][page needed]
> Rothermere and the Mail were also editorially sympathetic to Oswald Mosley and the British Union of Fascists.[48] Rothermere wrote an article titled "Hurrah for the Blackshirts" published in the Daily Mail on 15 January 1934, praising Mosley for his "sound, commonsense, Conservative doctrine",[49] and pointing out that: "Young men may join the British Union of Fascists by writing to the Headquarters, King's Road, Chelsea, London, S.W."[50]
People can down vote all they want, but doing so just confirms that the only way to satisfy them is for Twitter to pick sides and apply its rules evenly unfairly on behalf of one political party.
Kindness and caring isn't a finite resource that has to be "earned". If you begin caring for others by default, you might find that mindset disappears.
I hear you man, and that's a good point. It's just a lot easier said than done. I've been working on this stuff for over a year and it's just now starting to sink in, but barely.
