AI - What can it do for me today?

AI - What can it do for me today?
Photo by DeepMind / Unsplash

With the rapid advancements in artificial intelligence (AI) and large language models (LLMs), such as OpenAI's GPT-4, the potential applications of this technology in content creation and other industries are staggering. However, finding the ideal use case for LLMs has been a challenging process, and differentiating between artificial general intelligence (AGI) and LLMs can be confusing for some. In this blog, we'll explore what AI and LLMs are good at today, and how you can leverage their capabilities to optimize your workflow.

Artificial General Intelligence (AGI) vs. Large Language Models (LLM)

Artificial general intelligence (AGI) refers to the hypothetical AI that can perform any intellectual task a human can do, encompassing a wide range of cognitive abilities. On the other hand, LLMs are a type of AI model that has been trained on vast amounts of text data to generate human-like responses to prompts. While AGI remains a theoretical concept and is yet to be achieved, LLMs, in their current state, can assist users in generating ideas, summarizing large texts, and brainstorming, with the right guidance from the user.

AI and LLMs: What They're Good at Today

I've personally struggled with how to make LLMs work for me – keeping in mind that I haven't yet spent a lot of time with them – not being able to have it output something that doesn't just sound like "Oh this was written by ChatGPT.” I even complained about this on Twitter recently, generating a lot of feedback (including some automated bots powered by ChatGPT themselves):

What I learned (at least from the humans responding) is that the problem could be partially mine. It turns out that even though I thought I had a good handle on the difference between AGI and LLMs, I don't think I fully appreciated what that meant for how I, as a user, interacted with LLMs. Additionally, while "3.5" to "4.0" may seem like a small jump, I have to say that GPT4 is massively better at generating original-sounding text than 3.5 was...a fact I didn't really encounter until I ponied up the $20/month for ChatGPT Plus.

Furthermore, I've come to understand focusing on using LLMs for what they are good for. Ali Abdaal has a great video on this where he shares some ways he uses AI in his content creation workflow:

To generalize some key areas to keep in mind that "AI is good at today," think of things like:

  • Idea Generation: LLMs can help users come up with new ideas, storylines, or topics for their content. By providing a prompt, users can get a range of creative suggestions from the LLM, helping them overcome writer's block or explore new angles on a topic.
  • Summarizing Text: LLMs are particularly skilled at summarizing large volumes of text, enabling users to extract key insights and data from documents, reports, or research papers quickly and efficiently.
  • Drafting and Brainstorming: AI and LLMs can be used to kick-start the writing process by generating initial drafts or helping users brainstorm ideas, making it easier for them to develop and refine their content.

Even given that context, though, there is one more thing that is pretty important. You may have seen folks on the internet joking about being future “prompt engineers” ... but that's not that far from the truth! The right prompt can be critical in getting a LLM like ChatGPT to do what you want it to do.

Crafting the Perfect Prompt

A critical factor in getting the most out of LLMs is writing an effective prompt. By providing the LLM with the right context, users can significantly improve the quality of the generated content. To write a good prompt:

  • Be specific: Clearly define the topic, genre, or format you want the LLM to generate.
  • Provide context: Offer background information, examples, or guiding questions to help the LLM understand the desired output.
  • Set limits: Specify constraints such as word count, tone, or style to ensure the content stays on track.

There are several resources available online that offer sample prompts, such as OpenAI's prompt library, which provides examples of successful prompts that can help users get the most out of LLMs. Additionally, I've found these tools really useful:

  • Trickle Prompts, a warehouse of prompts folks have had success with
  • SecGPT: a prompt designed by Jason Haddix to aid with security-specific tasks
  • Awesome ChatGPT prompts: A GitHub repo with an ever-growing list of crowdsourced prompts
  • TypingMind: A better GPT UI and great prompt resource from one of my favorite indy developers, Tony Dinh

Other Applications of AI and LLMs Today

Beyond content creation, AI and LLMs are proving to be valuable tools in various other industries and applications:

  • Summarizing Recorded Calls: AI can transcribe and summarize long recorded calls, making it easier for users to review key points and action items. (there are dozens of startups building this type of tooling as we speak)
  • Text-to-Speech Conversion: LLMs can turn written text into natural-sounding speech, providing accessibility options for visually impaired users or creating engaging audio content for podcasts, audiobooks, or e-learning. (see
  • Machine Translation: AI and LLMs can facilitate real-time translation between languages, enabling seamless communication between people who speak different languages. (if you haven't tried the real-time Microsoft Translate app, I highly recommend you do!)
  • Sentiment Analysis: LLMs can be used to analyze customer feedback or social media content, allowing businesses to gauge customer sentiment and tailor their products, services, or messaging accordingly.

In conclusion, AI and LLMs have the potential to revolutionize the way we create content, communicate, and process information. By understanding the strengths and limitations of AI-generated content and LLMs, as well as mastering the art of crafting effective prompts, users can unlock the full potential of these technologies to enhance their workflows and optimize various tasks across industries. As AI and LLMs continue to develop and improve, we can expect even more groundbreaking applications to emerge, transforming our world in ways we have yet to imagine.