GPT-3 A Game-Changing but Raw Technology

January 30, 2023

Career / Business

Michel Schwartz recalled an eye-opening article that he read a while back about computers with artificial intelligence (AI). If computers became as intelligent as a well-trained CPA, doctor, or lawyer, the world would obviously go through massive change. Now, the business press was bubbling about the unbelievable capabilities of ChatGPT-3. Supposedly, this cloud-based AI tool was a big technological step forward.

What is GPT-3, and what does it mean for everyday people?

A picture containing person

Description automatically generated

GPT-3 Background

OpenAI is a research laboratory based in San Francisco specializing in the field of AI. The company recently made waves by releasing free online access to ChatGPT-3, a technology with the potential to revolutionize the field of AI. With its ability to generate highly realistic and coherent text based on simply “chatting” or typing in questions and requests, GPT-3 has a wide range of potential benefits and uses. (Chatting questions with GPT-3 tool equals ChatGPT-3.)

What’s It For?

GPT-3 is being used to generate text for a variety of applications, including creating content for websites, social media posts, and marketing materials. This tool helps people save time and resources by automating many forms of content creation with quality and consistency. Another potential use of GPT-3 is in the development of chatbots and other AI-powered systems for customer service and support. GPT-3 generates realistic and coherent responses to customer inquiries, helping improve customer service operations without human involvement.

Your Machine Dream Team

While this case for GPT-3 sounds a bit dry, the technology’s lifelike responsiveness and creativity across every topic are uncanny, even spooky. GPT-3 can be used in myriad applications, including translation, summarization, creative writing, financial research, and even writing complex computer code almost instantaneously. People playing around with the tool have, in a flash, produced interesting poems, fiction stories, and many other things you’d imagine you would need to pay a team of brilliant and creative people to create—immediately and at no cost.

Mind-Blowing Test Usage

The most eye-opening statement I saw so far about GPT-3 is from Hillel Fuld, a well-connected technology journalist and investor. Fuld wrote that this new technology has him “torn between concern and utter amazement, between fear and fascination. ChatGPT3 is a texting interface without a human on the other side. You are texting with a machine and it creates the answers! In real time! And you can ask ChatGPT anything, literally anything.”

To illustrate, Fuld had the AI write a sample business plan for a fruit store; it compiled it reliably within a minute. Then, a brilliant programmer friend of Fuld related how ChatGPT3 wrote a complex algorithm in 20 seconds, and it was the exact same one that the programmer recently invested 45 minutes into writing himself! The programmer added that this tool is the most significant technological step forward since the internet.

Big Deal—and Big Concerns

So clearly, we are talking about a very powerful tool, and we are just in the early stages. But while GPT-3 has great potential, it also raises a great number of risks, concerns, and unknowns. At this point, despite all the hype, we are a very long way from having all knowledge available at our fingertips in a safe and reliable way.

False Information

One of the main problems with GPT-3 is that the model could sometimes spit out misleading or false information. GPT-3 generates highly readable text, but it does not have the ability to understand or interpret the meanings. We live in an age in which dangerous rumors and falsehoods spread quickly via social media, and if AI recommends medicines or tax strategies, for example, it’s a big problem if they sound super-savvy but are actually wrong. When I asked the system to compare a SEP IRA and solo 401(k), GPT-3 answered smoothly but wrongly. A lack of trust in accuracy greatly limits the usefulness of the technology.

Spreading Falsehood, Hate, and Bias

GPT-3 raises serious ethical and social concerns too. AI tools may be used to make decisions that have significant impacts on individuals or communities but are guided by flawed or even manipulated data. For example, we’ve read much about anti-conservative bias at some tech companies, and there’s been a long problem of anti-Semitism at certain news outlets. Imagine that AI tools contain built-in anti-Semitic tendencies, leading law enforcement using the technology to speed up investigations and flag Jews unfairly! Systemic, unseen hate and discrimination is a much deeper problem in that case than what’s done on an individualist level.

Jobs Ramifications

Another question looming over AI is with regard to its potential to automate tasks that are currently performed by millions and billions of humans. While automation spreading into knowledge-based work may increase efficiency and reduce costs, it could also lead tup substantial job losses and drastic shifts of wealth to those with access to AI versus those who are left relying just on their meager brainpower. This particular risk is somewhat overblown, but it definitely can impact some or all of us in a very large way.

Governments Step In

AI also has major applications in warfare cybersecurity, technological development, and other things affecting national and international security. Governments around the world have and will regulate the use of AI. By shaping the development and use of AI through regulatory and funding mechanisms, governments can theoretically help ensure that this technology is used in a way that is safe, fair, and just. But government does not always do a great job, especially with big, complicated things. So they may smother the technology or actually introduce their own biases or abuses.

Dollar Power

Corporate control and development costs of AI will vary and affect the direction and use of this technology. ChatGPT-3 came to fruition thanks to a billion-dollar investment by Microsoft, and you can be sure they intend to get a return on investment. Corporate control of AI can lead to its use for the benefit of a few companies rather than the broader public interest while raising concerns about competition, transparency, and accountability. Businesses and governments will develop and control these game-changing tools as society is dragged along for the ride.

Now What?

Obviously, Hashem is in control, but to prepare for a world run by AI, we may choose to stay informed about developments in the field and learn new skills that could be required in a rapidly changing technological landscape. How it will all unfold I do not know, but like it or not, AI technology will probably make a massive impact on all of us in some way.

As a parting thought, note that the interesting if slightly creepy picture here was created instantaneously by Hotpot.ai, an AI tool. And most of this article was drafted in a few seconds using Chat-GPT3. Could you tell?