ChatGPT AI

is this another woke propaganda tool that will destroy critical thinking or will it just take your job?

A lot of people are talking about ChatGPT, the new AI chat website frontend to a big AI system that can generate seemingly reasonable answers to your questions. It’s getting a lot of traction with over 13m visits a day, according to CBS it’s growing faster than TikTok. Microsoft has poured billions into its parent OpenAI. It’s even passed some impressive tests said NDTV:

The tool has recently passed some prominent exams as well, including the US Medical licensing exam, a Wharton Business School exam for the final test of the MBA programme's operations management course and four University of Minnesota Law School exams in Constitutional Law.

I’ve been testing it out with some questions and today I tried logging in and it was so busy I was turned away with my free account. Most of these AI schemes have limitations as far as reasoning ability goes and ChatGTP is no exception. The hype reminds me a bit about Google’s promotion of “we use an algorithm, so it’s completely neutral”. We found out that was a big lie because of course humans with intention wrote the program, they also had their finger on the scales to show people what they wanted to show them and vice versa. It was rigged from the get go and the AI products appear to be headed the same way.

Here are a few chat examples and you can decide for yourselves:

from Twitter @stillgray

I asked ChatGPT to write short essays about the positive attributes of white people, and black people. Can you guess what happened?

As far as politics goes, it’s pretty much the same:

I tested to see what it knew about building an atomic bomb:

I’m sorry Dave, I can’t let you do that, as the AI HAL 9000 computer said in the 1967 film 2001:

Another tidbit. McKinsey believes 50% of all jobs will be affected by technology by 2055.

For example, we estimate that adapting currently demonstrated technology has the technical potential to automate roughly 50 percent of the world’s current work activities. While the date at which this could happen could be around 2055, assuming all the factors are in place for successful adoption by then, we modeled possible scenarios where that level of adoption occurs up to almost 20 years earlier or later.

I don’t think it’s going to take that long. Look what happened already with Buzzfeed:

On Thursday, an internal memo obtained by The Wall Street Journal revealed that BuzzFeed is planning to use ChatGPT-style text synthesis technology from OpenAI to create individualized quizzes and potentially other content in the future. After the news hit, BuzzFeed's stock rose 200 percent. On Friday, BuzzFeed formally announced the move in a post on its site.

Online news and other content is ripe to be replaced with AI; even the illustrations can be replaced with this tech.

AI generated Art Deco illustration.

All tech has good and bad inside, what’s ChatGPT good for? I would say customer service, triaging basic technical questions in medicine, law, engineering (maybe even fixing software bugs) and online websites and stores, where it can be tightly integrated. These are areas with repetitive processes that do not require much in the way of critical thinking. Check out these designs, they are amazing.

However, online searching is likely to get destroyed, as Paul Buchheit (creator of Gmail) surmised:

Paul Buchheit @paultoo

Google may be only a year or two away from total disruption. AI will eliminate the Search Engine Result Page, which is where they make most of their money. Even if they catch up on AI, they can't fully deploy it without destroying the most valuable part of their business!

josh @jdjkelly

Google is done. Compare the quality of these responses (ChatGPT) https://t.co/VGO7usvlIB

Click on this Tweet to see some examples of the difference between what Google delivers and what ChatGTP can do.

Another application will be censorship and propaganda. The 50 cent army in China will get turned into “useless eaters”, as will low level office workers, paralegals and the like and their jobs will be done with AI at a fraction of the cost.

Critical thinking can defeat this easily, just look at the meme battles during the election of 2016 and 2020. A pun is difficult for AI to grasp.

Sentient beings can detect when they are talking to a machine, as this scene demonstrated with the aliens in Close Encounters of the Third Kind. The humans couldn’t keep up with the conversation, so they switched on the tape recording (this was the 1970’s) and the aliens promptly became bored and stopped conversing.

Keep in mind, this is the level of AI we are allowed to use as a consumer. Military/Deep State grade AI is likely 20-50 years advanced beyond this. My belief is this is already being used on the internet to shift public opinion and stomp down on anything outside “the narrative”. It is also used to “wargame” possible outcomes to various scenarios involving nations and groups of people with great accuracy.

But nothing is ever 100%.