AI trained to do that job? Sure, yeah. LLM AI? Fuck no.
AI trained to do that job? Sure, yeah. LLM AI? Fuck no.
The article’s author mentioned that the problem is not limited to Samsung TVs - someone reported the issue on their phone.
The article does not mention a root cause, but I have a theory that it’s likely a malformed subtitle track. I tend to watch with subtitles on so I run into related issues every once in a while. Most of the time it’s one of two things:
The latter can have multiple effects depending on what format the subs are in, but most of the time it’s a missing end time, meaning that the subtitle stays on. However, some formats also have cues as to who the speaker is, and that comes with a start and end tag like in HTML. I suspect that in this case the end tag is either missing or misaligned in the syntax tree, causing this one line of dialogue to be displayed over and over when the player reaches other lines matching the cue for it, but that don’t get shown because the user has turned subtitles off.
As to why this is bleeding into other shows: I suspect it’s an issue with how the software clients are caching the subtitle files. This would also explain why going back into the episode that caused this fixes things, because it would reset the cached file. Which in turn brings me back to pointing the finger at Amazon, not Samsung, because Samsung would just be loading Amazon’s software client to play the video and subtitles.
Transparent vs translucent.
User-serviceable switches would be so nice…
Deep fried AI.
The patients or their families don’t even get the gift card, that goes to the hospital.
They do have a bpf sensor. It’s still shite, managing to periodically peg a CPU core on an idle system. They just lifted and shifted their legacy code into the bpf sensor, they don’t actually make good use of eBPF capabilities.
If the sensor was using eBPF (as any modern sensor on Linux should) then the faulty update would have made the sensor crash, but the system would still be stable. But CrowdStrike has a long history of using stupid forms of integration, so I wouldn’t put it past them to also load a kernel module that fucks things up unless it’s blacklisted in the bootloader. Fortunately that kind of recovery is, if not routine, at least well documented and standardized.
The London Underground is actually kind of a dumb use-case because it’s fixed infrastructure.
On the other hand it’s a perfect test bed, because there’s sufficient changes of direction and speed, and the fixed infrastructure lets you measure drift. Plus it being underground helps simulate GPS signal being weak or unavailable.
You can, but you’ll need to increase the microwave’s power accordingly.
It has implications on the effectiveness of VPNs on public networks.
A severe lack of imagination.
GM had at one point been working on an eCrate block for conversions, but they seem to have abandoned it.
That’s essentially Vivaldi now.
With a fast charger, owners will be able to get to 80 percent of battery life in just 30 minutes.
Well I don’t think I’ll want to kill 80% of my battery in half an hour.
They are also not a common fixture of garages.
Which means that your home then has increased heat loss because the garage door is open.
Some ICE cars also allow you to remotely pre-start
But you cannot do that in the garage (unless you like huffing exhaust fumes).
That’s LLM AI, but the type I’m talking about is the machine learning kind. I can envision a system that takes e.g. a sample’s test data and provides a summary, which is not far from what doctors do anyway. If you ever get a blood test’s results explained to you it’s “this value is high, which would be concerning except that this other value is not high, so you’re probably fine regarding X. However, I notice that this other value is low, and this can be an indicator of Y. I’m going to request a follow-up test regarding that.” Yes, I would trust an AI to give me that explanation, because those are very strict parameters to work with, and the input comes from a trusted source (lab results and medical training data) and not “Bob’s shrimping and hoola hoop dancing blog”.