The fight for AI: "Woke" vs "Based" AI, and whether AI will destroy "laptop class" workers.
Week of March 13, 2023
People have opinions about AI and it’s wild to see so many perspectives coming out of the woodwork. Friends, Noam Chomsky wrote an opinion piece for NYT about ChatGPT. It’s linked below. In this issue of my Weekly AI newsletter, I’m addressing an opinion piece about how AI will destroy laptop class workers, and the heated debate about “woke” vs “based” AI.
This is a response to an opinion piece I read on The Hill.
It’s not new news that jobs are, and will continue to be, impacted by AI. We saw a wave of job reductions from Microsoft issued in the same week they announced their 10 billion dollar investment in OpenAI, and I don’t think anyone believes that to be a coincidence.
It’s also not new news that new technology (or new applications of older technology) has forced humans to adjust since the first creation of tools. The manufacturing line replaced jobs. The invention of the computer replaced jobs. The invention of the combustible engine replaced the fire horses that pulled the old water carts, and before that, horses replaced humans who did the same job.
What makes AI a unique threat to employment are a few factors:
It is designed to mimic humans. It’s the first technology ever that has been created with the sole innovative purpose of mimicking what humans can do. Speaking. Listening. Smelling. (Yes, smelling.) Chores. You get it.
It scales quickly. Because the AI du jour (generative AI) is available as software and platforms, it can scale and be adopted very quickly. No need to buy/build/assemble/store in cases of generative AI. It’s just…available.
Its applications are great and many. AI is not just ChatGPT. AI has many branches, all of which will continue to see innovation over the next 10-20 years, leading to many use cases and various applications.
We’re already primed for integration. Technology is nearly everywhere, and a large portion of the global population has had access to technology. A sizeable amount of those people have had daily or near constant access to technology. The fifth industrial revolution is predicted to be where AI is integrated into our lives, and humans work harmoniously with the technology. Because we’ve been interfacing with technology regularly, we’re primed for this integration.
But here’s the thing…
I read this opinion article on The Hill, and while I can generally understand how this person drew their line of logic and why this individual made they claims they did, I don’t agree with their view, or how they shared their view.
I read it three times because I wanted to be sure that I didn’t miss anything, but from my assessment this article seems at best to be intentionally misleading and at worst to be sensational fear mongering. It shares a number of claims about how AI is going to “destroy the laptop-class” but offers little evidence to back it up, and there was little or no reasoning to explain the thinking that got them from point A to point B.
People who keep up with AI news may be able to read an opinion like this and decide whether it holds weight for them, but many people are not in the minutia of AI and this could cause them to be mislead.
My opinion on the article’s claims
Honestly, AI is going to impact jobs. It will. But it won’t replace all jobs, and it won’t replace them all at once. There are a number of things we have to consider when we think about how AI could replace humans in our jobs.
The first is human warmth to the idea of AI. Just because some people are rapidly adopting AI doesn’t mean everyone is rapidly adopting it. Companies the world over will take their time to bring AI in, and even when they do adopt the technology, they may not be comfortable with replacing a human with a machine.
The second are the resources necessary to find scalable, useful applications for AI. Computer chips, wires, etc. All of these things require natural resources for their creation. I’m sure innovation will occur in these spaces, but from a humanoid robot perspective, it would be tough to get the resources necessary to flood the planet with robots. Additionally, just because AI is AI doesn’t mean it’s good or useful AI. Poor application, inaccurate models or poorly built AI could lead companies to choose the old fashioned human approach for the near term, which reduces risk in many ways.
The third is “human-centric AI”. The more companies that adopt and prioritize responsible AI, the more likely it is that humans will be kept in the loop for any work where AI is designed, created, used, sold or deployed. This means a well-reasoned human needs to be able to see, understand and make adjustments when engaging with or overseeing the AI they are responsible for.
I think my biggest advice on this topic is the be really cautious about what you’re hearing and weigh it with other viewpoints and opinions. If a claim seems sensational and doesn’t offer evidence or reason to explain the claim, there is good reason to be skeptical. As always, please do your own research as well.
This week I was was hearing a lot of chatter related to “based” and “woke” AI. I wrote the article below, which include a primer, a mini evaluation, and some of my own thoughts.
"A new wave of heated conversation is taking place around whether the large language models (LLMs) for technology like ChatGPT and Bard should have restrictions aligning to a set of moral principles. This complex, philosophical, high-risk conversation has been oversimplified in a way that seemingly forces people to choose between two sides: "woke" AI and "based" AI.
Before we get into the nitty gritty, let's agree on a shared language.
Woke
The term "woke" was originally used to indicate that a person intentionally evaluated how their life impacted others, and made space for new people, viewpoints and opinions. Now, the term "woke" has been commandeered and has become synonymous with commonly held political views attributed to those who fall more on the liberal side of the political spectrum. The term is often used with a tone of disgust.
"Woke" AI is AI that is has restrictions in alignment with moral principles.
Rather than "woke" AI, since the term has been sullied, I'm going to use the term intentional AI.
Based
If you ask a friend who's deep into TikTok, a person who is "based" is someone who is being themselves and doesn't care what other people think. This term is also used in political circles and finds its latest application in AI.
"Based" AI is…"
Click here to read more.
One major challenge for individuals learning a new language is that they struggle to find people to practice with. Enter Speak. Speak is the first and only language app that leverages AI to help people get real conversational practice so they can learn a language through dynamic language application, rather than just memorization. The app has 66,510 app store reviews and a 4.8/5 star rating, and boasts more than one award.
Speak announced their $27M series B funding round in November, and in March they announced that their company is a launch partner with OpenAI, leveraging their automatic speech recognition AI. This is a natural partnership for the company, which previously open-sourced OpenAI’s “Whisper” models.
One thing that stood out to me about the company’s founders is that they spent more than a year reading research papers and learning about machine learning and LLMs. In the AI era where new companies are starting left and right, I can’t help but admire the research approach, and I’m looking forward to seeing how Speak leads the world in language learning.
Oversight:
Data & Privacy Concerns:
Opinion:
Book recommendation:
I regularly get asked how people can get into responsible AI, so here are some resources! I’ll keep adding to this list as I come across more information.
Responsible AI Institute is a nonprofit dedicated to helping organizations on their responsible AI journey. They provide awesome ways for their members to get connected through their Slack channel, get resources through their newsletter, and get invited to community events. Plus they’re a leader in responsible AI, so they’re a company to watch.
The Center for AI and Digital Policy offers policy clinics, and they look amazing. If you’re interested in AI policy, this might be for you! I’m hoping to apply for the Fall 2023 session.
Subscribe to my Substack if you want to receive this weekly update.
Connect with me on LinkedIn.
Do you need help on your responsible AI journey? Email me thepaigelord@gmail.com
Follow on TikTok, where I chat about responsible AI.