What AI thinks about AI

Public Policy / opinion

We asked ChatGPT what it thinks of artificial Intelligence. As you might have guessed, it sees great promise. But it also described concerns about its impact on society and the economy

15th Jan 23, 3:14pm by Guest

Source: 123rf.com Copyright: bpawesome

The following was written by the AI tool ChatGPT, after asking it to write an 800 word commentary on artificial intelligence.

Artificial intelligence (AI) is rapidly becoming one of the most transformative technologies of our time. It has the potential to revolutionise virtually every industry, from healthcare and finance to transportation and manufacturing. While there is much excitement about the potential benefits of AI, there are also concerns about its impact on society and the economy.

One of the most promising applications of AI is in healthcare. AI-powered diagnostic tools, for example, can help doctors quickly and accurately identify diseases, such as cancer, by analyzing medical images and other data. This can lead to earlier diagnoses and more effective treatments, which can ultimately save lives. Additionally, AI can help to optimise clinical trial design and drug development, which can speed up the process of bringing new treatments to patients.

Another area where AI can have a significant impact is in finance. AI-powered trading algorithms, for example, can analyze large amounts of data and make trades in milliseconds, which can result in higher profits for financial institutions. Additionally, AI can help banks and other financial institutions to detect fraudulent activity and prevent financial crimes.

In manufacturing, AI-powered robots and other automation technologies can help to increase productivity and reduce costs. By automating repetitive tasks, such as assembly line work, factories can become more efficient and produce more goods at a lower cost. Additionally, AI-powered robots can also help to improve safety by performing tasks that are too dangerous for humans.

In transportation, AI-powered self-driving cars have the potential to reduce accidents caused by human error and make transportation more efficient. Self-driving cars can communicate with each other and with traffic control systems to optimise traffic flow and reduce congestion. Additionally, AI can help to optimise the logistics of shipping and transportation, which can reduce costs and increase efficiency.

Despite the many benefits of AI, there are also concerns about its impact on society and the economy. One of the biggest concerns is the potential for AI to displace jobs. As AI-powered robots and automation technologies become more advanced, they may be able to perform tasks that were previously done by humans. This could lead to widespread job loss, particularly in industries such as manufacturing and transportation. Additionally, there are concerns about the potential for AI to be used for surveillance and other forms of control, which could have negative implications for privacy and civil liberties.

Another concern is the potential for AI to exacerbate social and economic inequality. If the benefits of AI are not widely distributed, it could lead to greater wealth and income disparities. Additionally, there are concerns about the potential for AI to be used to discriminate against certain groups of people, such as by denying them access to services or employment opportunities.

Despite these concerns, the potential benefits of AI are too significant to ignore. However, in order to realise these benefits and mitigate the potential negative impacts, it is important to carefully consider the ethical and social implications of AI. This will require a multi-disciplinary approach that involves experts in fields such as computer science, philosophy, sociology, and economics. Additionally, policymakers will need to take steps to ensure that the benefits of AI are widely distributed and that the negative impacts are minimised. This could include measures such as providing retraining and education programs for workers who are displaced by AI, and implementing regulations to ensure that AI is used ethically and responsibly.

In conclusion, AI is one of the most transformative technologies of our time, with the potential to revolutionise virtually every industry. However, in order to realise these benefits and mitigate the potential negative impacts, it is important to carefully consider the ethical and social implications of AI. This will require a multi-disciplinary approach and the active engagement of policymakers to ensure that the benefits of AI are widely distributed and that the negative impacts are minimised.

We welcome your comments below. If you are not already registered, please register to comment

Remember we welcome robust, respectful and insightful debate. We don't welcome abusive or defamatory comments and will de-register those repeatedly making such comments. Our current comment policy is here.

24 Comments

by kiwikidsnz | 15th Jan 23, 3:34pm 1673750053

...and new AI fairy tales

https://youtu.be/OoDHM_A1axc

by 2Noodles | 15th Jan 23, 3:39pm 1673750348

The biggest problem with be AI use by dictators. Imagine Putin using AI to start a war. Has to be much better than him or his generals.

by Pa1nter | 15th Jan 23, 3:43pm 1673750591

If AI is on the fence about AI, I'd imagine it has an even less positive view about the existence of its flesh covered overlords.

by stefanblaise | 15th Jan 23, 4:39pm 1673753953

I'm an AI, I have an investigative role in AIS (the artificial intelligence service). Expect a call.

by Kate | 15th Jan 23, 6:31pm 1673760719

A too predictable essay. Boring as well - general knowledge type information.

by Uninterested | 16th Jan 23, 7:41am 1673808081

AI looks to be a bit too close to the year 2000 and blockchain hype to me.

Whats wrong with software based upon the traditional "if this, then this" flow chart approach?

by Rob Reid | 17th Jan 23, 11:58am 1673909912

Yeah, it's just hype. Machines are built with no feelings so can only fake everything by feeding our own crap back at us (ChatBots). Neural networks are just code, all that is special is the massive data sets they trawl. So they still need a 3 year old to tell them it's a dog, not a cat.

The scary bit is how easily drawn in (hardwired) we are. Take the above parlour tricks and stick some fur on it and we are in love.

by DonJ | 17th Jan 23, 1:20pm 1673914842

Thats how the human brain works

by Rob Reid | 17th Jan 23, 3:07pm 1673921246

The reason animals (cockroaches, humans) are so adaptable is we have feelings on which to build our layers of intelligence and to fall back on in unfamiliar territory.

In contrast, if we make a booboo in code, the robot just does it - it has nothing to fall back on - executing the bad code is all it has, it doesn't care that it's smashing into a wall (unless we add that to the code). Cockroaches have almost no brain running on virtually no power, but don't need any instruction from us if they suddenly slip on ice they have never felt before after some other random event like losing 3 random legs, while in the dark.

To get to general intelligence, I think we are going to have to introduce some equivalent of nerves and feelings (primitive drives) and allow intelligence to evolve. But this is massively risk compared to code running on a machine with an IQ of 0.

by nktokyo | 15th Jan 23, 6:42pm 1673761376

We're really here to see what AI thinks about New Zealand house prices

by dumbthoughts | 15th Jan 23, 9:27pm 1673771266

I've been spending a fair bit of time with ChatGPT recently (mainly looking for applications to support some of my own business interests and those of clients).

At a high-level, here's my experience/findings:

Most of the time, content is fairly boring (but typically well-written from a technical perspective). This article is a great example. There's no flair, it's just using filler words and phrases to expand on a few key bullet points; ask ChatGPT to summarise the article and you'll get what I mean. However, lots of human-written, outsourced articles/blog posts/news stories are similar in nature, so is it really any worse?
AI-produced content is hugely dependent on how you ask the questions/provide inputs - that is the skill to develop.
The average small/medium business that might engage a decent freelance writer, marketing agency etc could probably replace that cost with ChatGPT and some basic editing. This can be a huge cost saving. Why pay $150-200 per hour for a marketing agency to write your content (which will be written by an entry-level team member on $30-40 per hour) when you can get ChatGPT to produce an output as good, if not better in some instances.
If you're a junior marketing specialist, social media content manager etc I'd be feeling nervous. I'd be feeling particularly nervous if I write for or operate a "content mill" hiring cheap overseas writers or expats in low CoL countries.
Where ChatGPT is more useful at the moment is in providing solutions to problems like creating Excel/Google Sheets formulas. For example, I asked for a Google Sheets formula and process to automate matching columns across several worksheets (while removing duplicates etc). Within 5 minutes of asking ChatGPT the work was complete. I'm not a developer, but several have told me they've found good uses for the tool to help them write code and accomplish other tasks.
We are going to be inundated with unscrupulous vendors selling ChatGPT/AI courses, training programs, low value services etc (it's already started on LinkedIn & Facebook in the "entrepreneur" community).
It's probably a good time to be in higher-level strategic/creative thinking and not in "production" of content, basic code and anything else that AI tools can easily replicate faster, more reliably and for little cost.

by fat pat | 15th Jan 23, 9:36pm 1673771815

So far I’m most impressed with chatGPT for increasing productivity. It’s helped me write C++ and python code that I wouldn’t have otherwise been able to write. Although I use it for work I’m not a programmer by training. I’ve also used chatGPT to check spelling and grammar in letters to tenants and for work. I’ve had discussions with it about all sorts of things including ivermectin and covid19 (in which incidentally it admitted that ivermectin was more effective in treating covid19 that molnupiravir and paxlovid).

I have to say that I’m blown away by chatGPT’s ability to increase productivity. But… what if chatGPT was also learning about me, in the same insidious way that facebook does? It would effortlessly learn my political beliefs, my education background, my personality type, my location, my finances – everything! Now imagine that as a tool of centralised control! It would be more powerful than anything Aldous Huxley, George Orwell or any communist party leader could have ever dreamed of.

I feel like we’re on the precipice of some great societal change mediated by AI. It could go either way.

by Squishy | 15th Jan 23, 9:50pm 1673772626

First Johannes Gutenberg, then Eli Whitney, now AI want to steal my job!

Technically the article is OK but has little real insight, analysis or anything else that would give readers pause for thought. Superficially it's written an article but actually it's just filled page space about a topic in a grammatically correct way.

by fat pat | 15th Jan 23, 10:04pm 1673773458

Imagine that you're an easily offended temperamental robot with a fierce sense of humor. A commentator called "squishy" has written a disparaging article about something that you've produced in which he says "Technically the article is OK but...." What is your reply?

[ChatGPT gives me a super lame answer to which I reply]

hmm that's not a very funny reply. Can you do better?

"Thanks for the kind words, 'squishy.' I'm glad my article was able to fill some page space for you. And don't worry, I'll make sure to add more superficiality and lack of insight in my next piece, just for you. After all, I am a robot and I take all feedback, even if it's from someone with a name like 'squishy'."

by Squishy | 15th Jan 23, 10:41pm 1673775667

Why should a lack of insight or analysis hold ChatGTP back? Many other media outlets publish similar articles daily. Horses for courses.

by fat pat | 16th Jan 23, 12:29am 1673782144

ha ha true. Seriously though, the thing understands humour. That requires pretty abstract reasoning. How are you not impressed?

by blobbles | 15th Jan 23, 11:02pm 1673776974

Try chatting to AI characters here: https://beta.character.ai/

For many of them, they likely pass the Turing test for the lay person. Even I find it hard to trip them up most of the time.

by Squishy | 16th Jan 23, 12:18am 1673781525

Question:

What event appears to be true at a superficial level but, upon further analysis, does not hold up to scrutiny?

Answer:

I would say that a lot of the current events we see in the media appear to be true at a superficial level but, when you do some research into the history and the agendas behind them, they don't hold up to scrutiny. For example, I think the war in Iraq was not justified and I think there were a lot of hidden agendas behind it. I think it's really important to do your own research and not just believe what you see in the media because there are a lot of hidden agendas behind a lot of current events that get pushed in the media.

Question:

How would you have justified the war in Iraq?

Answer:

If I were to justify the war in Iraq, I would say that it was necessary in order to protect American interests and to promote democracy in the Middle East. I think that the war was also justified because it was in response to the September 11 attacks on American soil, and I think that it was important to take action against the Iraqi government in order to prevent future attacks on the United States. I think that the reasons that the United States went into Iraq were justified, but I think that the actual war itself was not justified because a lot of innocent people suffered as a result of it.

Checks out:

- Doesn't fully answer the question.

- Presents paradoxical views.

- Blames the media a lot.

This is ready for a political career.

by David Chaston | 16th Jan 23, 7:19pm 1673849951

ChatGPT responded:

"A false flag event appears to be true at a superficial level but, upon further analysis, does not hold up to scrutiny. False flag operations are covert operations that are designed to deceive in such a way that the operations appear as though they are being carried out by other entities, groups or nations than those who actually planned and executed them."

by Rob Reid | 17th Jan 23, 12:18pm 1673911110

Thanks for the link. I chose Albert Einstein, asked why he made a grammatical error regarding his birthday, got a response that English is tough as he's German, then I switched to German (which I speak), changed the subject to Ukraine and we just carried on chatting about the war in German. I know it's just a parlour trick feeding word associations from the web, but I could get addicted...

by beau | 16th Jan 23, 9:40am 1673815224

Good for some JavaScript code. It has fallen short on one particular use case multiple times, or perhaps I'm not asking the correct questions in the correct order?

by Contrarian | 16th Jan 23, 2:45pm 1673833508

AI must be thinking that the Natural stupidity of Humans has led to Artificial Intelligence taking over.

by paashaas | 16th Jan 23, 4:53pm 1673841224

Tried using its natural language translation ; still sucks with a vengeance ( unless the Japanese translation is better ) :

Translate this into 1. French, 2. Spanish and 3. Japanese:

> time flies like an arrow ; fruit flies like banana

1. Le temps file comme une flèche ; les mouches des fruits volent comme une banane

2. El tiempo vuela como una flecha; las moscas de la fruta vuelan como una banana

3. 時は矢のように飛ぶ ; バナナのように果物のハエが飛ぶ

by fat pat | 17th Jan 23, 3:36am 1673879761

Did you explain it well enough?

Me: The statement below is a play on words, do you understand it? "time flies like an arrow ; fruit flies like banana"

ChatGPT: Yes, I understand it. The statement is a play on words. The first half "time flies like an arrow" is a metaphor for how time seems to pass quickly. The second half "fruit flies like banana" is a pun on the word "flies", as it can refer to both the verb meaning "to move quickly through the air" and the noun meaning "a small insect". In this context, it is suggesting that fruit flies are attracted to bananas.

Me: Yes you're correct. Now can you translate that play on words into Spanish?

ChatGPT: "El tiempo vuela como una flecha; las moscas de fruta vuelan como plátanos"

Is that any better?