I mean, fair point. But it sure does smell like the original commenter's kids have been brainwashed. "Booing" a car or having a hateful attitude towards a car without any external influence? I actually don't think the Cybertruck looks good - at all. But all my kids are amazed by it. And we haven't spoken to them at all about Elon.
Totally, what kid would boo something without being preconditioned to hate it? Humans can never separate concepts in there heads. Either you’re good or you’re evil. Have news for you all, we are all both. Not saying I support that guys behavior, but he’s done some impressive stuff to further mankind.
I’ve always shared this skepticism. First, crypto can be a poor store of value when its price is highly volatile and most goods and services are still priced in currencies like the U.S. dollar. Second, some crypto networks can make transfers slow or expensive, which limits their usefulness for everyday transactions. That said, it varies by asset and network—some are designed to make transfers faster or cheaper.
I’m also uneasy about the investment case. When you own crypto, you have a reason to want others to buy it too. That can create a cycle where demand pushes the price up, and the rising price attracts more buyers. That happens in other markets too, but productive businesses can also generate revenue and cash flow that help support their value. Many crypto assets don’t have that same kind of foundation, although some are tied to specific services or network activity.
Regular currencies have a more obvious role in everyday life (pay taxes, settle debts, and buy things). Their value is influenced by factors like monetary policy, confidence in the issuer, and the strength of the underlying economy. That doesn’t make them stable or valuable in the same way as productive assets, but it does give them a practical purpose that many crypto assets don’t have.
Very sad. I will always remember my first exposure to Patrick Stewart - Comptons Interactive Encyclopedia. Such a fun time at the beginning of my computer adventures in the 90s.
It’s really pathetic. I would think even the lowest quality LLMs could figure out tool use like this. It has to figure out the tool and most of the time, it’s stated in the command. Like remind me or put this in notes, so why can’t it do this more easily. I’m just having trouble figuring out why. I’m curious if they still have tons of rules in the background in a way that is muddying the pure LLM experience.
I’ve spent a lot of time testing the new Siri. Overall I like it and believe this will become a very powerful tool but definitely think it needs work.
- Example: “Siri for the lights that are already on in the family room, can you put those lights to 50% brightness?” -> All the lights turn on incorrectly. This is a two step process. Get information from Home and then execute command with that info.
- Example: “Remind me to call back Joe at 5pm.” -> Siri oddly has trouble with this. Asking me what time after I give this command. Is it not hearing full command?
- Example: It seems to ask me to clarify the name of something when that name is readily available in recent personal context. Like recent 5 messages. It should be able to know some of this stuff as I’m most likely operating on recent stuff.
It seems to be hampered in some way. Gemini should be able to handle that tool call really easily.
Other things are great. Like I have my checking account and routing number buried in Notes somewhere. When I’m in safari and ask it to autofill with the routing number from notes. It does it. Amazing.
I’m so bullish on LLMs and my personal context but there are some kinks here that hamper performance so far.
I have counter evidence of this. My wife with absolutely zero skills in computers was at the terminal doing things that Claude was telling her to do and generated some impressive tools. One was a tool to help her organize her day. It involved scripts, PDF generation, printing. She did it all without even asking me. Maybe it fails in some cases but I'm not sure the anecdote above is the average experience.
I have to say. Nice work. I spent years wanting to build something like this to integrate to these systems. I never did but have spent years thinking about this. So, nice work and really appreciate your motivation.
I keep going back to the debate with myself. Do I give my now-young kids 500K in an ETF in 14 years or do I send them to college? And the debate is not about how they will spend it (I can hold on to it too), it’s about whether to spend the money or just pass on the wealth to them. People have given me tons of reasons why college is good for them - independence, etc etc. I can find a way to give them independence without spending 500k. I haven’t settled the debate but articles like this are not helping the college path.
Funny. He foreshadowed this in a recent interview. Saying that he may fall out of touch with evolving approaches and if any of the frontier labs would have him, he’d be interested.
Does anyone know of any HR departments actually using LLMs for scoring, selection, extraction, classification or any real use cases? I'm curious to hear about it and how they are using it.
How many proprietary use cases truly need pre-training or even fine-tuning as opposed to RAG approach? And at what point does it make sense to pre-train/fine tune? Curious.
I still use AirPods for listening but if I’m ever taking a call, I always use EarPods (USB-C). The microphone quality is multiple times better and that’s important to me. Especially for work. It only took me a few times to hear other people will AirPods to be tainted. It just seems unprofessional now because of how bad it sounds.
Mostly same story. Tinkered for hours with Windows 3.1 floppy disks. Reinstalling OS’s all the time because I’d break stuff or I’d just want a fresh slate. I loved pushing the boundaries. In my 30’s I slowed down with the tinkering because of life (kids, work). I thought I lost the ability to tinker. But recently at 42, I bought a MacBook for the sole purpose of tinkering on the couch at the end of the day (basically after being on computer the whole day, I didn’t want to be in office anymore). And slowly, it’s coming back. I’m playing with new things, learning about Neural Networks, learning about Softare Defined Radio, installing tons of random libraries and tools to test that out. It’s coming back. Keep pushing on it and hopefully it returns for you too!
Has anyone tried to use a laptop at night? It’s pretty hard without lit keys. Maybe this one has some super reflective letters so that the screen lights them up.
My advice is a little different. It’s make the life of your boss insanely easy. Similar in nature to post but slightly different optimization function. Don’t over communicate, communicate just the right amount. Anticipate questions. Don’t create any friction for them and be really helpful. Some of my people will anticipate things and be proactive. I love that and I constantly push to get them promoted.
When I read the publications (the ACM magazine), I swear sometimes the content feels LLM generated. Does anyone else get that impression? In general, I'm not very impressed with the content (I'm used to WIRED, btw).
The way I think of it (might be wrong) but basically a model that has similar sensors to humans (eyes, ears) and has action-oriented outputs with some objective function (a goal to optimize against). I think autopilot is the closest to world models in that they have eyes, they have ability to interact with the world (go different directions) and see the response.
This is a very smart idea. I couldn't turn my Ring Alarm off and I was on the same Wifi connection as the system. In retrospect, it would be quite smart to switch over to local network.
Exactly. This was my point. Televisions can upconvert from 720p to 4k. In the same sense, the machine learning model would fill in the waveform and mimic a high powered mic. It can do this at the connection point (iPhone / computer).
Isn’t there a whole bunch of dependency here related to prompting and methodology that would significantly impact overall performance? My gut instinct is that there are many many ways to architect this around the LLMs and each might yield different levels of accuracy. What do others think?
Edit: In reading more, I guess this is meant to be a dumb benchmark to monitor through time. Maybe that’s the aim here instead of viability as an auto close tool.