ChatGPT Drinks More Water Than You

Mar, 2024

$.00 PER COPY

👋 This is The Pipette – a monthly newsletter for product builders with links so good their ideas warrant a reply, a forward, or even a discussion in real life.

This isn’t necessarily an AI-focused newsletter, but it’s been a hard topic to ignore — generative AI is moving at a blinding pace, and has been in major news cycles since widespread adoption started with ChatGPT. For better or for worse, more AI tools are being adopted and lots of us are figuring out how to adapt.

Our approach so far has been: to automate repetitive tasks, get some initial ideas on a topic we’re not as familiar with, and run calculations/scenarios through conversation.

We’re also finding that we can get more done in most cases, and sometimes it can help us go above and beyond in areas that we would not have otherwise. For example, using GitHub copilot makes commenting code & documenting so fast that it’s fun. It’s also easy to add more robust error-handling, test cases, etc.

In terms of it coming for our jobs, that hasn’t happened. If anything, by being able to create robust experiences, we’re just saving ourselves from heartache in the future. That’s not to say that other areas won’t be affected though. And there are plenty of legitimate ethical & privacy concerns with how large language models source their data, especially for artists and writers. Like all technology, it can be misused, and it’s ultimately on us all to support companies & efforts that drive it in the right direction.

To wrap up, here are some AI articles resources we’ve found thought-provoking. You’ll also find some more general links we like further down.

AI assistants boost productivity but paradoxically risk human deskilling – AI may be the great equalizer: it can boost employee performance for the least-skilled workers, while it had less of a benefit for the high-skilled ones, even causing deskilling by increasing dependency on it.
ChatGPT drinks a water bottle every 5-50 prompts – dig into the environmental implications of the processing that LLM data centers require. It definitely made us want to get better at prompting!
The emerging field of prompt engineering is the way to do just that. Just like AI, it’s been moving fast. If you’re interested in learning, check out this Prompt Engineering Roadmap.