kenneal - lagger wrote: ↑17 Jan 2023, 19:20
That's one of the problems of continuous evaluation over examinations.
Maybe/hopefully we'll see the balance change back towards more 'old style' tests and exams...
Appreciate they don't suit everybody, but they make it much harder to fake what you do/don't know...
We already have "open book" examinations, why not "open link" ? AI is a tool to be used by a human, currently, maybe exams should reflect how well a candidate can use available tools.
Catweazle wrote: ↑17 Jan 2023, 21:16
We already have "open book" examinations, why not "open link" ? AI is a tool to be used by a human, currently, maybe exams should reflect how well a candidate can use available tools.
So if you don't know your subject but are good at computing you should get a qualification and job in anything you fancy? Doesn't bode well for innovation in British industry or education.
Catweazle wrote: ↑17 Jan 2023, 21:16
We already have "open book" examinations, why not "open link" ? AI is a tool to be used by a human, currently, maybe exams should reflect how well a candidate can use available tools.
So if you don't know your subject but are good at computing you should get a qualification and job in anything you fancy? Doesn't bode well for innovation in British industry or education.
I agree, but tools come in many shapes. Does a checkout operator need to do maths in his head ? A Uber driver need to memorise the roadmap instead of using GPS satnav to calculate a route ?
Google has now thrown it's hat into the ring with it's AI offering, Bard. It'll be interesting to see how this shapes up against chatGPT. I've been playing around with DALL.E and it really can produce some quite good art, especially if you're a fan of pop art. I can see this putting paid to companies like moonpig.com and funckypidgeon.com, as you can make greetings cards etc for next to nothing.....
"Roose says that before it was deleted, the chatbot was writing a list of destructive acts it could imagine doing, including hacking into computers and spreading propaganda and misinformation.
After a few more questions, Roose succeeds in getting it to repeat its darkest fantasies. Once again, the message is deleted before the chatbot can complete it. This time, though, Roose says its answer included manufacturing a deadly virus and making people kill each other."
I just asked chatGPT to tell me about my business.
It was partly accurate, but completely wrong in many ways that it shouldn't have been. The 'embellished' descriptions of the wrong stuff I don't get, it didn't have to guess at things as factual accounts of the place are all over the web including on several large websites.
It came across as trying too hard and just making stuff up when there was no need to. It's a puppy trying too hard to be liked.
Yes, learning as I go. It's interesting. The next step in people not using their brains, that's rapidly becoming apparent. We'll all be like my 14yo nephew soon, utterly useless at everything that doesn't involve a mobile phone. I'm betting even switching on a pc would be beyond him, and all his friends. Young people are truly stuffed.
I have seen reports it has just been banned in China. Or, according to the Grauniad work arounds to get access have been closed.
China will focus on its own version.
Hope it is better than the old version.
Plato, a chatbot released by Baidu in 2021, drew unfavourable comparisons on social media after it failed to match up to the new US-created entrant, according to a translation by ChinaTalk. In one widely shared example, Plato became fixated on saying “3+5=5”
There are loads of new websites appearing, full of seemingly plausible but sometimes inaccurate content. You can spot the inconsistencies through the articles. Example, I looked up the specs of some steel, the webpage delivered referred to the steel as "14%" chromium, "nearly 10%" chromium and various other "high chromium" all on the same page, as well as mixing in various myths about stainless and plain carbon steel and loads of keywords relating to applications.
This, combined with the inability of Google to deliver relevant pages, is a serious impediment to finding accurate information on the web. Apparently some experienced IT people are now adding the word "Reddit" to any Google search now, in an attempt to find content generated by humans.