I'm using chatGPT to generate social media posts for a business. I just give it variables that need to be mentioned, what tone I want the post to be (optimistic, somber, etc.), and any key words I want included.
I'm also using it to help with my studies. My favourite instruction is 'explain it like I'm 12' - because it uses simplified language to explain concepts that my textbooks made seem impossibly complicated.
I'm aware of both the limitations (extreme bias, timescale of existing information isn't current, etc.) and the concerns from institutions, but I'm loving it!
Would be nice if you could ever get onto the site. Literally EVERY time I try it says "oops our systems are a bit busy at the moment. Try again later". Effin' useless.
I must be the only person in the world that's not had a play with it yet.
My manager suggested taking a look at it. I work with a bit of C++ for numerical simulations, python for data visualization, and MySQL for the boring work of keeping databases running.
Have not played with the new toy, but I think Google/Bing work fine as long as you know what you’re looking for. When I have problems, I’ve found it takes 90% of effort to make the right question/search. Not sure how the new toy will take the job of making the right question for me. The above mentioned pieces of code are aimed at very specific tasks in my team’s workflow, so not sure…how the AI can read my mind or learn about how my team works.
Then…data security. If you have an interesting job, don’t use the thing:
My impression is that VCs are on a rush to find bagholders to buy their equity on useless AI solutions before next crisis hits and the valuation of useless solutions matches their real potential. That’s why mass media is pushing the narrative of AI being awesome at everything.
Of course, there’s potential in the technology but nothing related to your favorite newspaper columnist that 1 week writes about AI, week before climate change, and on two weeks about war in Ukraine. Please, don’t read or listen to your favorite columnist/opinion maker
This is interesting, just yesterday my brother said he was thinking of getting a subscription because subscribers get priority access and he can never get on it. Whereas I didn't even know subscription existed because I've never had a problem.
This right here - I like the interactive nature of it because oftentimes I don't know what I should be asking / looking for. For me, that's the selling point of AI over traditional search engines.
I was going to ask it to introduce it self to the forum, but seems the site is down at the moment... How's it ever going to take over the world with an attitude like that...
Not only does the paid service get you priority in getting access, more importantly it buys you access to the GPT4 engine, which gives considerably better results.
Ok it does know who we are... but maybe not in all our nuances
The English Forum Switzerland is a website and online community that caters to English speakers living in Switzerland, as well as those who are interested in the country and its culture. The site is used by a diverse group of people, including expatriates, international students, business professionals, and tourists. The forum provides a platform for its users to discuss a wide range of topics related to Switzerland, including travel, food, culture, politics, and language. It also provides information and advice on practical matters such as housing, healthcare, and employment.
Much like any other source of information in history.
I'm sure everyone here has stories of factual mistakes from books, teachers, professors, news, professionals, academics, google....
For quick information (although I have seen some mistakes as well), it is a welcome change now that google is SEM/SEO'd into uselessness. Even reddit searches are losing their value.
I wanted to open a new thread about this topic and now gladly I’ve seen this existing thread.
The topic is truely highly interesting and involves, beside absolute good ideas, even hight risk when miss used. And this will happen earlier or later for sure. From who, we will see.
Auto-GPT: The new tool that gives you a peek into the AI future | This World
In fact it will transform society (not chatGPT itself but the LLM generative AI) and possibly take huge swathes of employment meaning we will all exist on UBI or that society will descend into chaos.
It makes some mistakes occasionally now but in the main it generates novel
Content which is good enough. Customer service, first line contact centres, help desks, journalism are the first under the hammer. More will follow.
Ideas like ‘it won’t take my job because I’m just so irreplaceable’ are completely missing the point. It’s already cleverer than you are, or will be, and it will get better exponentially.