it seems like a good chance that despite thinking their status page was completely independent of cloudfront, enough of the internet is dependent on cloudfront now that they're simply wrong about the status page's independence.
It would be next to impossible for anyone without insider knowledge to prove that to be the case.
Secondly, benchmarks are public data, and these models are trained on such large amounts of it that it would be impractical to ensure that some benchmark data is not part of the training set. And even if it's not, it would be safe to assume that engineers building these models would test their performance on all kinds of benchmarks, and tweak them accordingly. This happens all the time in other industries as well.
So the pelican riding a bicycle test is interesting, but it's not a performance indicator at this point.
can you point me to a useful example of this? i see websites including ai-generated summaries all the time, but i've yet to see one that is actually useful and it seems like the product being sold here is simply "ai", not the summary itself - that is, companies and product managers are under pressure to implement some sort of AI, and sticking summaries in places is a way for them to fill that requirement and be able to say "yes, we have AI in our product"
I sometimes get contracts, NDAs, or terms and conditions which normally I would automatically accept because they are low stakes and I don't have time to read them. At best I would skim them.
Now I pass them through an LLM and ask them to point out interesting, unconventional, or surprising things, and to summarize the document in a few bullet points. They're quite good at this, and I am can use what I discover later in my relationship with the counterparty in various ways.
I also use it to "summarize" a large log output and point out the interesting bits that are relevant to my inquiry.
Another use case is meeting notes. I use fireflies.ai for some of my meetings and the summaries are decent.
I guess summarization might not be the right word for all the cases, but it deals with going through the hay stack to find the needle.
Do you go through the haystack yourself first, find the needle, and then use that to validate your hypothesis that the AI is good at accomplishing that task (because it usually finds the same needle)? If not, how do you know they're good at the task?
My own experience using LLMs is that we frequently disagree about which points are crucial and which can be omitted from a summary.
It depends on how much time I have, and how important the task is. I've been surprised and I've been disappointed.
One particular time I was wrestling with a CI/CD issue. I could not for the life of me figure it out. The logs were cryptic and there was a lot of them. In desperation I pasted the 10 or so pages of raw logs into ChatGPT and asked it to see if it can spot the problem. It have me three potential things to look at, and the first one was it.
By directing my attention it saved me a lot of time.
At the same time, I've seen it fail. I recently pasted about 10 meetings worth of conversation notes and asked it to summarize what one person said. It came back with garbage, mixed a bunch of things up, and in general did not come up with anything useful.
In some middle-of-the road cases, what you said mirrors my experience: we disagree what is notable and what is not. Still, this is a net positive. I take the stuff it gives me, discard the things I disagree on, and at least I have a partial summary. I generally check everything it spits out against the original and ask it to cite the original sources, so I don't end up with hallucinated facts. It's less time than writing up a summary myself, and it's the kind of work that I find more enjoyable than typing summaries.
Still, the hit to miss ration is good enough and the time savings on the hits are impressive so I continue to use it in various situations where I need a summary or I need it to direct my attention to something.
I really don't see how it can save you time if you have to summarize the same source for yourself every time in order to learn whether the AI did a good job in this particular case.
There is a difference between creating a summary and double-checking if a summary is correct. It's kind of like coding something, or doing a code review. It does save me time, but that might have to do with how I think and what I enjoy doing. Everyone is different.
for your first one, if you're just feeding docs into a chatbot prompt and asking for a summary, i think that matches what the article would call a "chatbot product" rather than a summarization product.
fireflies.ai is interesting though, that's more what i was looking for. i've used the meeting summary tool in google meet before and it was hilariously bad, it's good to hear that there are some companies out there having success with this product type.
I guess you’re right re chatbot for summaries. I was thinking about the use case and not the whole integrated product experience.
For example, for code gen I use agents like Claude Code, one-shot interfaces like Codex tasks, and chatbots like the generic ChatGPT. It depends on the task at hand, how much time I have, whether I am on the phone or on a laptop, and my mood. It’s all code gen though.
We built a system that uses summaries of video clips to build a shorts video against a screenplay. Customer was an events company. So think 15 minute wedding highlights video that has all of the important parts to it, bride arrival, ring exchange, kiss the bride, first dance, drunken uncle etc
you can click the share icon (the two-way branch icon, it doesn't look like apple's share icon) under the image it generates to share the conversation.
i'm curious if the clock image it was giving you was the same one it was giving me
No, my clock was an old style one, to be put on a shelf. But at least it had a "13" proudly right above the "12" :)
This reminds me my kids when they were in kindergarden and were bringing home their art that needed extra explanation to realize what it was. But they were very proud!
Copilot completions in vscode are pretty great, and i think a lot of people are happy with that.
in general i agree with you, adding an AI chat window to an app that isn't an AI chat app is almost always a detriment. but i think it's shortsighted to assume there won't be other important use cases for AI, and we're in the experimentation phase right now where companies are trying to learn what that looks like. it's just unfortunate that there's so much incentive for apps to frame their AI chat as the best new thing ever and you should really use it, instead of introducing it more subtly.
>Why should apps have access to a user's SMS / RCS?
can you imagine the outrage from all the exact same people who are currently outraged about develeloper verification if google said they were cutting off any third-party app access to SMS/RCS?
the difficult part of the highways is the interchanges, not the straight shots between interchanges. and iirc, tesla didn't do interchanges at the time people were criticizing them for only doing the easiest part of self-driving.
>In diseased tissue samples of people with chronic illnesses (IBD [6], Dementia [7], heart disease [8]), microplastic prevalence is significantly higher than healthy tissue.
this is very much not the same thing as "microplastics cause chronic illness", even though it's worded in a way that clearly wants to make you think that.
The precautionary principle is bad epistemology and shouldn’t be used to argue in opposition to anything. If we’re considering actions to ban or reduce microplastics it should be backed by a reasoned explanation for why we should do so.
And where does it point toward? Other some untenable position such as "ban all plastics", which may very well produce more harm?
The discourse around microplastics is pretty wild. The sport is finding them in random places, often at parts-per-billion or parts-per-trillion levels that we don't really use to look for most other substances. And the implication is essentially "progress bad" or "consumerism bad". No clear evidence of human harm, no realistic policy prescriptions - so what do we expect to happen, exactly? This it not a case of corporate greed or deception.
Our bodies also contain a fair amount of sand. Probably at levels higher than parts-per-billion. Is it bad? Sometimes! Where does the precautionary principle lead us on that?
Make the plastic manufacturers own the external costs by requiring they fund proper disposal sites/messaging, if only to start making up for all the bullshit propaganda about recycling that's greatly exacerbated the problem.
> Make the plastic manufacturers own the external costs by requiring they fund proper disposal sites/messaging,
I think you fail to understand the reality of the problem.
In western countries, plastic "trash" is not really the problem. It's highly visible and it would always be nice to reduce it of course.
The majority of uncontained environmental microplastics comes from vehicle tires and clothing/textiles. Clothing and other textiles (e.g., carpets) being the biggest source, more than 1/3rd. After that it's probably building materials, paints, machinery and factory parts, etc.
Disposal sites and messaging will not do anything. You can be a perfectly compliant goodly consumer who dutifully puts their old clothes in the trash and pays the disposal fees for their old tires or rides busses. You'd still be contributing enormously to environmental microplastic load.
All natural fiber clothes, cycle everywhere, don't wear sneakers or other kind of plastic or synthetic rubber shoes, don't have synthetic carpets or drapes, don't paint your house, etc... now you're starting to get somewhere.
But the machinery required for you to stay alive, moving goods and services around, pumping your water, people going to work to keep your electricity on, package your food, etc... all still pumping out microplastics.
Disposal and messaging just won't cut it. And without a bunch of astounding and vanishingly unlikely breakthroughs, getting rid of microplastics from the top 4-5 sources will make net zero CO2 look like a walk in the park. Therefore we have to accept microplastics at enormous scale and work with that. Not to say we shouldn't attempt to reduce it where possible of course we should, but it won't be reduced to insignificant. So I think what needs to be done is well funded research into the effects of existing and new types of plastics, and into new materials and techniques for cleanup or containment. That way we have a chance to discover and limit or ban the worst of the worst before they can become too pervasive.
As far as reduction goes, possibly some small incentives to avoid plastics in consumer items (clothes, carpets, etc) might help. The messaging really can not be the same idiotic and counterproductive alarmism and blame and guilt campaigns led by wealthy private jet and mega yacht owning billionaires of the climate change debacle. Just gently make people aware they could look for natural fiber clothes, perhaps modest and commensurate added costs on plastics manufacturers to fund this research and containment, etc.
Why not start with the large sources that you can personally control?
> One 2023 study published in the journal Environmental Science & Technology found that microwaving plastic food containers releases more than 2 billion nanoplastics (smaller microplastics) and 4 million microplastics for every square centimeter of the container.
"Ban all plastics" is a strawman that will not happen and no mainstream opinion is suggesting. But there is a wide spectrum of possibilities between "ban all plastics" and "do nothing".
A principal concern is ingestion of microplastics via food packaging, utensils, cookware, etc. There are non-plastic substitutions available for many of these items, and a precautionary approach would be to regulate to require them, where it is economically feasible, until such time as the effects of microplastic ingestion are better understood.
I don't think there's any more room for not considering underestimating the importance of beginning to start the process-of mulling over the conceptualisation of starting to worry.
It's saying that there could be a link and a link has been found on mice. That and the fact that the human body is not supposed to be running partially on plastic should trigger some actions.
Exactly the same argument could be (and unfortunately, has been) made about vaccines. For all we know, microplastics could be making us healthier. I don't believe that's true, and I'm avoiding plastic where I can, but the fact that we have it doesn't necessarily mean it's bad.
Not really. With vaccines you can have controlled groups (vaccinated or not) and study the benefits versus risks. This is done before any vaccine is introduced to the public, any vaccine showing not enough benefits versus risks is discarded.
We do not have this luxuary with plastic. Controlled groups are not possible, asking that risks inherent to micro-plastics in human body are proven is therefore impossible.
To compare back to vaccines, it would be like vaccinating very single human, with a vaccine which is not tested and which purpose is unknown. Then asking someone to scientifically prove that this vaccine causes health problems as a condition to stop systematically vaccinating every single human.
it took me a minute to parse that sentence also. They are saying that health tissue makes up < 100% of the body, but that microplastics can be found in a full 100% of the body (healthy and non-healthy). Therefore microplastic prevalence > healthy tissue. It's saying that there is no part of the body that isn't impacted
I think it's saying there's more microplastics in unhealthy than healthy tissue is all, your interpretation is technically possible but phrasing it that way would be so misleading as to basically be lying.
The reason more microplastics in unhealthy tissue doesn't necessarily mean microplastics cause unhealthy tissue is that unhealthy tissue would be worse at removing substances irrespective of whether the substances cause the harm.
reply