Why we need to stop criticising AI and accept it!

Last Updated on March 22, 2023 at 9:02 am

Before the end of 2023, AI will play a huge role in almost all spheres of life. At freefincal, we have been obsessed with creating a CHATGPT-based AI bot to answer reader questions. More about this is below. First, let us discuss why we need to stop criticising AI and accept it!

Every time someone talks about how wonderful CHATGPT is, there is always this familiar criticism: (1) It is not perfect; (2) AI can never replace human beings; (3) AI is no match for human creativity and ingenuity and so on.

While these are all true, they are not particularly useful.

(1) Of course, AI is far from perfect, but this is partially because it is incredibly hard to build robust training data (and we speak from experience). Automated solutions to create an input data set are full of useless information, and manual efforts take too much effort. That said, I will not bet against it becoming better.

There seems to be some critical data set size that is not too small to result in poor responses and too large to result in unreliable responses. The freefincal bot is currently being tested by a few readers. We will know more soon.

(2) Yes, AI cannot replace a high IQ or creative individual. However, most of our tasks are mundane and based on rules and patterns. An AI can replicate this – maybe not everytime, but enough to consider it an assistant if not a replacement.

(3) The harsh truth is most of us are mediocre. And AI-driven solutions or content or entertainment are “good enough” for most people. Naturally, there need to be checks in place for important areas.

Take, for example, “evergreen content” in finance, fitness, travel, history, science etc. The rules in such areas are rarely going to change. Today I can whip up a 300-500 word AI-authored article on the “importance of a budget” or “basics of retirement planning” in seconds.

Any writer would tell you that the headline is everything. And today, we have plugins that can generate alternate headlines in seconds. And most of them are not just good; they are, “oh, why didn’t I think of that” level excellent.

We have AI to write code and poetry, draw excellent pictures, speech-to-text and vice versa, create videos with human-looking presenters and generate pictures of life-like models doing whatever we want them to do. It feels like everyone suddenly has access to a powerful “special effects” department for various tasks, and it is brilliant.

At freefincal, we are trying to use AI to solve simple user questions that would lead them to read more of our articles and use more of our products. It is a matter of time before AI offers serious, contextual financial advice.

We strongly believe in robo-advisory. Our current tool can either be used to DIY a financial plan, or advisors can create their own robo avatar by modifying assumptions suitably. If an AI model can be trained on the robo tool, users can create a financial plan over a chat window! It will take some time to get here, but we are that far off.

Many establishments already use an AI-trained bot to answer potential customer questions and support. This will improve in the short term. The potential is mind-boggling.

Will AI replace jobs? It will modify job roles for sure. There will always be jobs to go around.

The freefincal bot is based on about 1000 articles and is intended to help new readers and potential product users clarify the way forward. Initial testing suggests that it is neither terrible nor excellent. A small group of readers is currently testing it. Their feedback will decide the way forward. This is a screenshot of a beta test.

Update: We released a bot from another provider to the Facebook group Asan Ideas for wealth, and the results are encouraging enough to pursue it. However, most users are asking for specific investment advice from the bot. So more training is necessary to dissuade this properly.

Currently, we are testing five different bot creation channels, and once we choose one and are confident about its responses, it will be added to the site.

Meanwhile, in his blog, Bill Gates writes, The Age of AI has begun. “Artificial intelligence is as revolutionary as mobile phones and the Internet”.

As regards AI killing jobs, this is the smartest take on the subject by Dr Jim Fan.

Stop saying: AI will replace humans. Start saying: humans who know how to use AI at work will replace those who don’t.

Artificial intelligence is amazing, but to exploit it requires human intelligence.
Do read this Twitter thread (if you have not already) about an ongoing experiment to earn income online using CHATGPT-4

The possibilities are mind-boggling. And they say CHATGPT-5 is coming soon!
Worrying about AI, hating it or criticising it is a waste of time. Our only goal should be, “how can we improve our skills and, if possible, our income using it?”, “How can we push ourselves to ask better questions to the model?”

A small fraction of us will push ourselves hard to get better and better responses from AI and grow our skills, knowledge and even wisdom. That is the group I would like to belong to.

In summary, I believe criticism of AI, although correct, is non-representative and way too harsh. It should not distract us from the reality that AI is the future. The path will not be easy, with giant pitfalls to navigate, and as with everything, we must live and learn. The future is exciting.