Using GPT to generate an app's GUI on the fly. Basically, the design of the app is transient and created dynamically based on the exact need of the moment. https://mastodon.social/@stroughtonsmith/110193697691243762

LLM prompting with a visual programming language https://arxiv.org/abs/2304.08103

The end sample mentions injection of expert domain knowledge, and calls out psychology and mock interviews. Interviews? Sure. Psych I’m not sure.. I can hardly trust a human to understand my meanings and intent therapeutically, and the context build up is part of building session value up. That seems quite hard to stuff into a small sliding window.

I don't know if I want to laugh of cry 😂. This is done just by duct-taping a bunch of APIs and GPT-4 prompts together. Maybe the future of programming will be more English than code? https://chameleon-llm.github.io/. Paper (https://arxiv.org/abs/2304.09842) and Code (https://github.com/lupantech/chameleon-llm)

Been thinking about the infra needed for human-machine teaming, I think a lot of knowledge from the field of business process management can be reimagined, just with powerful natural language capabilities. • Natural language becomes the API interface between humans and machines. • No need to formally describe the workflow. Just talk to it and let the LLM create/modify/delete these. • These workflows are essentially Turing-complete, distributed program where the job "processors" are people and bots. For eg: BPMN already supported branching, sequential processing, parallel processing etc. • Open-source Zapier-equivalent systems can be a good starting point for these.

‘Godfather of AI’ Geoffrey Hinton quits Google and warns over dangers of misinformation” - maybe old news, but I haven’t seen it mentioned here... https://www.theguardian.com/technology/2023/may/02/geoffrey-hinton-godfather-of-ai-quits-google-warns-dangers-of-machine-learning

Perspective on AI from a music producer (He reminds us of Napster and concludes that networks, like Paramount+, Apple Music, Spotify, etc. will simply side-step artists and music labels altogether)

AI autocompletion like Copilot was just the beginning - we're going to be getting more agents trying to automate much of the process of software engineering. The future of coding is starting to look like management. https://github.com/smol-ai/developer/

• humans have cognitive limits, • e.g. the Magical Number Seven, Plus or Minus Two • https://en.wikipedia.org/wiki/The_Magical_Number_Seven,_Plus_or_Minus_Two • does GPT have similar cognitive limits? • is this a Golden Number? a universal truth? a universal constant? • is it necessary to arrange information in a hierarchy? • or, does GPT’s perfect memory change this pattern?

This might fall under "thinking-together", but I'm throwing this here as it specifically relates to AI. For context, I've been working to integrate an LLM into an existing application. If "prompt engineering" is part of the future of coding, here are some challenges to expect: • Non determinism: I can repeat a prompt and get back drastically different results, in both content and format. • Capabilities: I feel like I'm constantly probing to discover the capabilities and limits of the model. Every interaction is similar to being presented with a blank canvas: What should I ask? What can I ask? Is there something that I could ask that I am forgetting about? Can I reword my prompt slightly and (potentially) get better results? This leads to a lot of experimentation. • Expectations: We are tricked into believing the LLM comprehends what we are saying, when it's really just a giant prediction table. Then we are disappointed when it gives less-than-satisfactory replies.

Compare this to programming against a traditional API: • On in the input side, an API constrains the vocabulary. With an LLM, everything is fair game. • On the output side, I can guess (or learn) the effect of a given API call. With an LLM, it's all probabilistic.

Curious about others experiences here, ways to reason about these models, techniques for overcoming, etc...

Is anybody using gpt to autogenerate blog posts or other educational media?

Grokking

In 2021, researchers made a striking discovery while training a series of tiny models on toy tasks . They found a set of models that suddenly flipped from memorizing their training data to correctly generalizing on unseen inputs after training for much longer. This phenomenon – where generalization seems to happen abruptly and long after fitting the training data – is called grokking and has sparked a flurry of interest .

https://pair.withgoogle.com/explorables/grokking

An interesting take on AI risk: https://metamoderna.org/what-can-stop-the-ai-apocalypse-grammar-yes-only-grammar/ The vision exposed in this article is that humans with their institutional superstructures (bureaucracies, markets, corporations, ...) and AIs (plural) will/should form an ecosystem in which all players coevolve, competing and collaborating at the same time. I don't think that any of the players are ready for this, but in the long run, this is where we could be heading.

Hard to pin down what's most dystopian about this https://openai.com/blog/teaching-with-ai

Doug Lenat, inventor of Cyc, has passed away

Apart from things like the langchain and hugging face blogs, does anyone have any good resources for the latest tools and techniques for LLM prompt engineering? I need to bone up, this week.

One of the more successful AI code generation tasks I've experienced so far (for non-trivial code) is creating Python code from pseudocode (found on wikipedia.) Has anyone had experience with using pseudocode representation and LLM's?

I don’t know how I missed it in the announcement, but when you use an LLM chat cell in Mathematica, it sends all the cells above it as context. I realized that was happening when the LLM happened to refer to code I’d written in one of its answers. Now I just straight up ask it “Why?” when something surprising happens. I haven’t made a single non-chat notebook since I figured this out. 🌈

Did a technical assignment for an interview this week, and made extensive use of ChatGPT4 to great success. It's to the point now that if I get an error message I don't understand, or get to a step where I'm unclear what to do next, I'll try ChatGPT4 before I will try Google, because with ChatGPT4 it already has the context for the problem, and gives me an answer specific to my context. Very different programming experience.

i feel like python is very limiting for prompt engg/agents. anyone else feel this way? maybe we need more expressive languages/dsls/languages built with editor in mind (like smalltalk).

A.I. is simply pattern matching - a summary of a talk I attended … https://guitarvydas.github.io/2023/10/27/PEO-Generative-AI.html

Hey 🙂 I am looking for inspiration in building frontends with LLMs. I am looking for companies, which have LLMs (their own or wrappers around GPT) deployed in customer-facing applications (B2B and B2C) at some level of scale. Something like Sudowrite and Jasper. Also if you know some impressive solobuilders, feel free to name drop them 🙂 I would appreciate any link or name dropped in the thread 😉 Best Nicolay

I keep directing people to this talk: https://maggieappleton.com/forest-talk @Maggie Appleton explores it better than anyone else

I don't know much about LLMs, so I'm purely asking out of ignorance. AFAIK Open AI has a powerful LLM system in large part due to large compute power. Is there a future where small business / people can compete with mega-corps in the LLM space, or will we be renting GPT (or similar) until end the of time?

IMG_0891.png