or Nash Equilibrium. If EU wants to get away from MS and GOP's flakey tribalism, they have to agree to all (or most) moving at once, and accept a learning curve.
I'd go with Nash Equilibrium here. DVORK is not inherently better than QWERTY for typing speed, although it is less prone to causing RSI. The big downside is the learning curve and making shortcuts like Ctrl-V harder to press.
To that one user: "Tough shit. You're getting and using the updated software or you can find yourself a new job." Assuming you have that power, of course.
User: The annual report is nit ready because IT forced a software update and it broke the report generation. I tried to warn them but they did it anyway...
No browser is going to bring back the good old days.
I want Ludwig Plutonium, AFU, Kibo, Serdar Argjic and the Zumabots, Ted Frank, and Joel Furr vs. The Green Card Lawyers...
These organizations are offloading their technology deficiencies onto the poor and incapable user.
I, as a seasoned professional that has seen every trick email that has ever been sent, cannot determine if certain messages are malicious or not. The users will never be able to do it themselves. And punishing them, whether that's reprimand, scolding, mocking, making them resit training, is NOT the right way to do it.
Exactly. IT orgs should put defenses in place to catch the majority of fishing emsils, using sender info,AI to recognize patterns etc. You won’t catch everything but can greatly reduce the chan crew s they get through.
You do know that you wouldn't have to sign up for such a scheme, right? There's no gun to your head?
Yes, this is understood...however, that's an indictment against the category. Games are optional entertainment products, so the problem is that *any* legislation of entertainment products can come under the guise of "it's optional, you can just not-buy it". Agreed; it's not food or medicine, so the industry benefits from a lower level of necessity in that respect.
However, "it's optional" doesn't mean it's not worth regulating. Gambling is "optional", yet slot machines have such strict regulatory compliance requirements to meet, that they commonly exceed the level of scrutiny given to voting machines and MRI hardware. The gun-to-head argument would stipulate that since casinos are way easier to avoid than hospitals or banks, slot machines don't need regulation.
When video games started running into areas that consumer protection laws typically cover, we started needing to assess the appropriate level of regulation. For decades, games were one-time, permanent purchases - people with Nintendo cartridges and Playstation games on disc can play them today, 20-40 years after release, and neither Nintendo nor Sony have a practical means of saying "...you can't play that anymore". Over time, games got tied to accounts, and there were some consumer benefits - no more scratched discs or lost media or shelf space issues, plus additional content and bug fixes after release.
These were great, until they weren't. Those accused of cheating saw account bans, which meant they couldn't play *other, paid-for games* anymore, irrespective of whether the accusations were true. Companies would take money for games, but then shut down the servers needed to play them saying "it costs us too much", but then would sue fans who said, "...we'll take it from here, then". Of course, there was the whole "surprise mechanics" thing, where games were literally-gambling and they tried to argue that it wasn't.
So yes, there's no gun-to-head requirement for people to sign up for these services. Stadia proved that. However, the problem is when such systems are implemented as a requirement, for artificial means. If GTA6 was "Stadia Exclusive", it's not like GTA6 can't run on consoles or PCs, it's purely a predatory business model for a game that inherently doesn't require it. It's "pay us the same amount as before, but we can claw your access to the game back, but you can't claw your money back, because this benefits us". Now sure, purely rational human beings would refuse to buy it until they released it with traditional distribution mechanisms, but 'vote with your wallet' is exceedingly difficult to make work at scale.
Even if you want to argue this is acceptable for video games, it's a short hop to business software, where such a model has harmed many small businesses with things like Quickbooks pulling this crap and forcing businesses to go from $300 every few years to $2500 annually. "It's optional!! they can print out 20 years of financial data and use paper and a pencil!!11" - how far does that argument go, exactly? that's the problem with the gun-to-head argument - if *only* unavoidable needs receive regulation, then everything that is "optional" becomes a potential means of exploitation. There needs to be a balance, of course, but somehow I don't think that video game publishers are unreasonably impacted because the government tells them "you have to remove DRM from your products if you're not selling them anymore, and you can't sue your customers for fixing things in your products if they paid for it, either".
...if I need to connect it to an account, and/or it needs internet access to function, then it's a guarantee that 1.) the conversations will be stored in Amazon's meat grinder, 2.) the user won't be able to delete them (just make them invisible for themselves), and 3.) it's going to turn into a subscription fee.
Anthropic is likely training its models with your data. If your data is cruel then the models can become cruel as well.
Considering they likely used credit,
not to review techical documentation to identify design issues
But this is exactly where AI can actually be useful — so long as you don't depend on it to identify all of the design issues. It can review mountains of information and find things in it you might not have caught. You just can't assume that it will catch everything it "should" — that is, what it should be capable of catching is not necessarily what you are capable of catching.
Yes, and it requires a multi-layerd approach to how they system raises issues. One simple prompt pass is not enough, however a multi-tiered approach can increase teh success rate. It'll never catch everything, and neither will a human, the tool is merely a way to remove much of the grunt work searching though documents so teh analyst can focus on what is important and make it easier ot find where things are referenced.
Meanwhile, that periodic report has requirements, so you're going to have to ask it to write the report, then to review it to see if it met the requirements, then to review its own statements about whether it successfully met each requirement, because it won't do any of that sanity checking without being told to do so.
There are two different types of reports to develop - one that validates meeting requirements and one that is merely reporting results or periodic required reporting. The former is challenging, the later a lot less so, and reducing the time spent on the later frees up time for the former. Of course, many such reports do not need AI since they merely compile data, and that is easy to program in any of a number of tools. AI just adds hype and market cap...
While I agree that AI is not truly deterministic, that doesn't mean you will get widely divergent results from teh same data.
Widely? No. Divergent? Yes. Run the same query multiple times and you'll get different answers. The more input you give, the more subtle the differences are likely to be, but there's always a chance it will go completely off its nut and follow some line of "reasoning" that goes someplace utterly unrelated to what you asked for. This is less likely in special-purpose models, but never impossible.
Of course, which is why for special purpose models this can work; especially since you can train it to be more accurate once you have enough real life data. This tool is not likely using the commercial ChatGpt but a bespoke setup.
Tailoring the LLM to taht can save a lot of time for an analyst/engineer/operator by identifying and collecting teh information needed for analysis and reporting
It can't, though, and for two reasons. One, a person is required to check off on the data. Two, LLMs don't faithfully represent data. They ingest it as tokens and then what they crap out is influenced by random numbers produced internally as part of the token selection process. That means a person cannot certify that the data was faithfully reproduced without checking literally every value, because that's simply not how LLMs work.
You can potentially address this with a hybrid approach, where the LLM uses a tool to collect and manipulate the data, but this is shifting the problem to whether the LLM used the tool correctly. It might be more obvious when it doesn't, though, so it might still be a win.
What LLMs seem best at when it comes to text processing, and they are still not infallible, is summarization. But you can't trust its negatives, so it's only useful for identifying other possible positives. That's not worthless, it can help you catch things, but you can't trust it to do the parts you'd most want it to do.
While I agree that AI is not truly deterministic, that doesn't mean you will get widely divergent results from teh same data. There is some run to run variant, but in my experience, working with very specific criteria, it tends to be in the ranking order and not a failure to identify relationships. That can be addressed in a number of ways. While positive hits are easier, negative ones can be checked in a number of ways to increase the confidence level that they truly are negative, in the sense they are not relevant to the tested data set. To me, one of teh keys is to work with a well defined set of criteria, something I think is possible with tehy type of information they are dealing with, especially since much of the input data will be well structured and in a consistent format.
A human must always be the final arbitrator, and needs to realize the AI is not fool proof and you can't accept everything at face value.
I wouldn't want them to be relying on AI to do reviews though. Maybe two AI, looking only for issues and reporting them back, like you can have AI look for security issues with computer code. And a proper review done by a human.
I was think more along the lines of "Here is the periodic report on X" or "Here all the similar events and corrective actions..." to basically act like a research librarian; not to review techical documentation to identify design issues.
Academy was a great Trek show. It captured the spirit of Star Trek and showed us how Starfleet helps young people develop into officers, while also giving us some new and interesting ideas. The characters were interesting, the legacy stuff wasn't obnoxious and actually added to what came before (especially for The Doctor after what B'Lanna did to him), and the overall writing and presentation overall was good.
Sorry Ami, but I'm gonna have to give a bit of a counter here.
Starfleet Academy was referenced in TNG and DS9 as a prestigious institution, only accepting the best. Wesley Crusher and Nog were both shown as really needing to be at the top of their game in order to be accepted into the academy. In Academy...pretty much everything was "find yourself and who you feel you really are, and the team will work around that".
Meanwhile, the setup for the story was that Holly Hunter left Starfleet because she was upset about a particular mother and son (Caleb) she separated when she was a judge, so she left, and now Starfleet needs Her In Particular to be a professor at Starfleet Academy (really, a whole universe full of people and we need one person who wasn't a teacher and resigned?), and she goes to find Caleb In Particular to join Starfleet because he's good at everything, in exchange for helping him find his mother...except that the "find my mother" plot thread is barely there, and not even Caleb seems to care for most of the season.
As the show gets going, Starfleet Academy (the institution) takes place on a ship that can just land and be a building, depending on what's necessary...and they go on a bunch of super dangerous missions that really require seasoned officers to do, but they have a bunch of cadets doing stuff they clearly shouldn't be doing...for, some reason. And they go out in space to do it instead of doing it all on the holodeck where it's far safer for...some reason. And, somehow, they're almost never in class in this college, except for one episode where we watch them be in debate class. Okay, fine, but...no engineering classes? No tactical training? No indication that they're actually learning to do the things we've seen Starfleet officers doing in the field?
Most of the show centered around the interpersonal drama, and I'll even grant some leniency here because it was pitched as having a "Dawson's Creek / Harry Potter" sort of coming-of-age focus. uhm...okay, fine...but quite a number of episodes had so much of that sort of focus, that it barely needed to be "Star Trek" to do it. They brought in the 'war college' to create some allusion to the Harry Potter houses, but do we spend a meaningful amount of time with sympathetic characters in The War College? No, we don't. Do we have any meaningful thought put behind the possibility that perhaps there's a need for a military-focused part of Starfleet in a universe that has physics-defying pirates (the last episode requiring an impossible feat of travel and resource acquisition), the way that Enterprise had the MACO's or the Elite Force games had the Hazard Team?
Finally, at least for my personal opinion, I found absolutely every single character insufferable for one reason or another. Holly Hunter is a captain-slash-professor who tells her leadership to get bent, and is shocked when the cadets have authority issues? I forget her name, but the Jem Hadar lady was generally annoying and useless, and somehow I got the feeling that half the cadets could have taken her in hand-to-hand combat. Caleb was "the key to everything", Genesis was every popular girl trope simultaneously, Ja'den was the living embodiment of 'not-Klingon'. The writers were clearly going for 'child-like' with SAM, but ended up at 'child-ish' instead. Tig Notaro was Tig Notaro playing Tig Notaro. You get the point.
Please note that I got through my whole list of exceptions without complaining one bit about "wokeness" or "politics in Star Trek" or anything related to it.
Flock doesn't solve violent crime, it might provide evidence for a policeman to solve violent crime.
It's not evil to solve violent crime. You are getting over excited dear.
However Flock could be used for evil. And there is a consumer backlash because of that, it's the wrong product.
And that is the conundrum. If there were real efective data protections to ensure private data was secure from any cop wanting to do a search and searches required teh same effort as those in the real world, people might have fewer issues with it. Collecting massive data bases of people is inherently bad beacuse it will be missued, as events have already shown.
More interesting is why is there is so much violent crime in your culture? If you are an American, you appear to have a violent culture, perhaps the most violent in history,
The US culture certainly grew out of violent struggles to expand its territory across the continent, and no doubt shaped its current one to a degree; and has it roots in European culture. Most violent in history? There are plenty of cultures that can easily compete for that title, from European ones to Asian to African and everywhere between. Europeans, for example, routinely massacred or enslaved defeated opponents in their history.
the only one to have dropped a nuke.
The moral of that story is don't keep fighting a war you can't win, and when it is obvious you are going to lose, surrender.
<< WAIT >>
