I love hearing what people are making with ChatGPT and the other AI chatbots. Tons of interesting and innovative ideas. It’s fun to use new tech tools! Embrace your creativity!

Are these chatbots way overhyped? Yes.

Are people talking about this trend as if the sentient AI robot revolution is already here and humans are about to go extinct? Unfortunately, and of course, yes.

Are people using these tools for bad or just banal purposes? Almost certainly.

But…

But that doesn’t mean the tools themselves are bad. We can make chatbots, LLMs, and AI in general better, safer, fairer, more prosocial, and more sustainable.

We can and we must. Because if we don’t work on this now, don’t build in the safeguards and figure out protections and remedies, we know harm will occur.

tl;dr AI is great. Overhyped but full of potential—for good and for bad. Let’s develop more of it but do it responsibly.
@tiffanycli #AI tools aren't inherently bad but unfortunately the people utilizing them are. They're being used to stampede over artist copyrights, generate and flood the market with tons of generic/formulaic fiction, and enrich a handful of techbro billionaires at the expense of working artists. The last is true of most new tech, but I think the current suite of AI tools is an extreme example. I can't agree with supporting these tools until the misuse is extricated from the tech.

@tiffanycli I mean, just watch #MidJourney's CEO casually disavow interest in the art copyright issue like an all-white brunch squad that "doesn't get into politics"... as he rakes in millions while trampling small time artists into the dust with a copyright-eating behemoth. Starts around minute 14:

https://www.youtube.com/watch?v=Sqa8Zo2XWc4

Artificial Intelligence: Last Week Tonight with John Oliver (HBO)

YouTube

@tiffanycli

i wouldn't even say overhyped.

i believe/fear we don't even grasp the impact #ai (incl. #chatgpt) is going to have.

@tiffanycli
I spent a bunch of time last fall playing with Stable Diffusion. It was a wonderful creative experience, and got me thinking about the world in different ways. I hope that we can see more open source tools, with training on public domain datasets moving forward
Deep Blue computer beats world chess champion – archive, 1996

12 February 1996: IBM supercomputer makes chess history by beating Garry Kasparov, the world’s best chess player

The Guardian
@tiffanycli What we did not expect was that they were bad sports...
https://www.youtube.com/watch?v=L-77oMd0USw&ab_channel=SkyNewsAustralia
Chess-playing robot breaks boy's finger during tournament in Russia

YouTube
@tiffanycli we experimented with ChatGPT to create position descriptions. They do a great job of getting about 80% of the way there. For non-HR specialists, such as Managers, I can see the potential for these Chatbots to significantly reduce both the work, and the stress, of creating position descriptions.
@tiffanycli I also get the issues with these chatbots. But there is a strain of criticism that feels like people laughing at, say, a young giraffe learning to walk. Kinda punching down almost.

@tiffanycli

time will tell whether #chatgpt or other AI will either boost or kill human mankind's creativity.

it will depend on the rules we define...

🤔

@mad yes! the rules and regulations will be critical
@tiffanycli Gosh you're an upbeat lady

@tiffanycli @mmasnick

1. A super silly use of the technology, but I wrote a short iOS Shortcuts shortcut which asks chatGPT to write me a rhyme using the current time (which Shortcuts adds into the prompt). Just for fun… (see attached pic).

2. An iOS Shortcuts shortcut that runs when you invoke Siri (e.g. by holding down the side button on your iPhone or Apple Watch) and say “Hello Einstein”. Prompts you for voice input of a question and reads the answer to you — to have a not-dumb-Siri!

@leoncowle @mmasnick Maybe silly, but also fun and creative! If it makes you smile, why not?