The world is changing fast with new developments and breakthroughs every week. From recent banking crises to ChatGPT’s plugin store, it’s hard to keep up!
Waitlists, waitlists everywhere!
Whenever there’s a new and exciting AI technology on the horizon, instead of simply being able to use it, we always find ourselves on a waitlist. While waiting for access to new AI technologies can be frustrating, it’s a reality that we must accept. However, by being patient, we may ultimately reap the benefits of these innovations and gain a competitive edge in our industries.
First ChatGPT and then Microsoft’s Bing GPT based chat companion, and now there’s the latest plugin store. Seeing all the hype on the Internet I found the website and… joined the waitlist. Let’s see how much time it takes to get access.
I have high hopes for the plugin store, as it presents a great opportunity to improve productivity. If I can prompt the model to do my social media cross-posting, or to check key websites and send me an email with the summary and links, it would be an amazing tool to have at my disposal.
Moreover, I’m also excited about the possibility of prompting my hosted files and getting results based on them for my website or app. From the quick look at documentation at Introduction – OpenAI API, all of those should be possible.
ChatGPT 3.5 – impressions
I’ve been using ChatGPT couple times for the past 3 months. I intended to write what did I use it for, but instead I run this prompt to ChatGPT: Can you check all my chats and summarize what I’ve been using ChatGPT for?
Certainly! Based on our conversation history, you have been using ChatGPT to:
- Ask for advice on video editing software
- Compare and contrast different video editing software programs
- Request assistance in writing and improving your Substack newsletter
- Ask for suggestions on how to phrase and structure your writing
- Discuss the latest developments in AI technology, specifically related to ChatGPT and OpenAI’s API
- Seek help in improving your writing and grammar
- Ask for recommendations on books to read
- Engage in casual conversation and exchange pleasantries
It completely forgot I was able to have Python code created, even though I’m not a developer (yet!).
This is great start and I’m planning on using it a lot, not only to do things better and faster, but also to keep myself up to date with how the transformers transform the world. Many people I enjoy listening to and trust have said this GPT like models will increase productivity by orders of magnitude and I want to check myself if that’s the case.
There was an interesting take on this from Chamath Palihapitiya on The All-In Podcast E122 this week. He discussed how companies like Cognizant have great incentive to use it to increase the output of their development teams in the developer for hire model.
Chamath: What I would tell you, Jason, to answer the white collar question is, I think there are a handful of companies you need to look at exclusively because they will be the first ones to really figure out how to displace human labor, and that is TCS, so Tata consulting services, Accenture, Cognizant. These are all the folks that do coding for higher work at scale. I think Accenture has something like 750,000 employees, so the incentive to sort of squeeze Opex to create better utilization rates to increase profitability is quite obvious, it always has been. They will be the first people to figure out how to use these tools at scale before the law firms or the accounting firms or any of those folks even sort of try to figure out how to displace white-collar labor. I think it’s going to be the coding jobs and it’s going to be the coding-for-hire jobs that companies like Accenture and TCS.
JCal: So those business processing do for other people developer kind of folks they’re going to need half as many people, 25% as many people?
Chamath: We’re going to find out the efficient Frontier.
There also have been counter arguments from David Sacks and David Friedberg, if you’re interested, have a look at the podcast:
My Weekly Media Roundup
- https://youtu.be/qQ544sWC8ZQ The All-In Podcast (link above)
- https://ir.tesla.com/press-release/tesla-vehicle-production-deliveries-and-date-financial-results-webcast-first-quarter-2023 Tesla Vehicle Production & Deliveries and Date for Financial Results & Webcast for First Quarter 2023 | Tesla Investor Relations – also many tweets coming from me on the topic.
- Eliezer Yudkowsky: Dangers of AI and the End of Human Civilization | Lex Fridman Podcast
https://youtu.be/AaTRHFaaPG8 - Stephen Wolfram on AI’s rapid progress & the “Post-Knowledge Work Era” | E1711. Interview with a person behind Wolfram Mathematica – a tool I used extensively around 2006 or a bit later – during my university studies.
https://youtu.be/F5tXWmCJ_wo
Wrap up
To wrap up, I want to thank whoever has reached this point (if anyone). I’m most active on Twitter, where I’ll be posting short messages as I test out the Twitter Blue subscription to gain some traction for this idea of sharing my thoughts on business and technology. A list of all my channels follows: