AI could replace the equivalent of 300 million jobs – will yours be one of them? Here’s how to prepare.

Last year, many of us spent time thinking about the problem of AI bias, which was carefully outlined by one of the authors of the famous “Coded Bias”. Netflix Documentary. Now it’s yet another boost Generative AI While the popularity is here to stay, talk of job replacement is back in play.

Namely, one of the most verbose reports on how AI could potentially automate (or, as many fear, replace people in the jobs they deserve) Goldman SachsAbout which was spread intensely under various alarmist headlines 300 million potentially changed jobs worldwide.

In particular, some reported data suggest that 18% work worldwide are likely to be computerized, and the effects on more developed economies may be worse than on emerging countries, for example.

Unsurprisingly, the recent boom in generative AI has coincided with several consecutive waves of layoffs in the online tech industry, making only a small amount of panic more understandable in numerous discussions on the web.

Related: 3 Principles of Building Anti-Bias AI

However, the report itself suggests that the so-called “exposure to automation” does not in itself imply the elimination or elimination of human-involved jobs. More importantly, many non-white-collar occupations are not even vulnerable to negative impacts.

On a larger scale, according to some experts, the ability to manage next-gen AI tech will be crucial for professionals, rather than being rendered redundant by Chat GPT-like solutions anytime soon. As Ingrid Verschuren, head of data strategy for Dow Jones said“Humans are the real “machines” that drive AI.”

Facing the reality behind the hype

So, according to Goldman Sachs estimates, up to approx 25% of all work can be managed by AI Absolutely in the coming years. But what does this mean for, for example, a legal department specialist, copywriter or motion designer? To be honest, not so much.

A friend of mine, who runs a video production studio, tested it AI solutions Generating images for some time and scraping creative inspiration from machine learning algorithms has been a tedious journey all along. The default imagery is often somewhat generic (and often dark for that matter), so their design team hasn’t been able to actually implement the newly acquired AI-powered assistance to a significant extent.

Meanwhile, the latest trend is playing out in editorial departments ChatGPT Some news personality quizzes and seeing non-true results have also proven that the issue of veracity is the weakest point of generative AI.

And so many false stories are given, and how easily generative AI tools Being persuasive (eg write stuff with non-existent facts, if it’s given in the assigned request), I highly doubt an inexperienced, yet hungry paralegal is qualified enough to go along with their legal advice. But leave it to them. Software equivalent yet.

Will the future hold our fears?

While the current state of generative AI is apparently not as advanced as its founders believe, some job market predictions for 2024 seem too pessimistic for that matter. Of course, technologies are likely to have a significant impact on our workforce in one way or another in the coming decade. So how can we prepare?

Here are some focus points that entrepreneurs can keep in mind:

Don’t rush the cut-off

No matter where you are in business, Generative AI doesn’t have the skills and competencies to replace any qualified experts on your team in the current state.

More importantly, even further AI progress Come on, you’ll probably need your team to manage the new software (ie explain exactly what needs to be done, then review the results) to get the best results.

Some of the most vivid examples include code reviews/tweaks, AI-generated script editing, accounting and engineering project re-checks, and physical exams/prescription reviews in medicine, but the list is virtually endless.

Related: History has shown what happens to companies that shy away from new tech, so why are so many people afraid of generative AI?

Check your facts

While we leave the media and celebrities worrying about the potential negative effects of complex deep fakes, made possible by the introduction of generative AI upgrades, using ChatGPT or similar tools to find information is a very difficult business.

As the training of algorithms evolves, the risks of being completely misled will certainly decrease, but chances are we won’t be able to trust AI-generated text/images in the near future.

This aspect will be of primary importance in editorial newsrooms, law firms and political offices, however, any calculations provided by Advanced Machine Learning Algorithms At least in selected data sets, re-examination will also be required.

In particular, the amount of time and operational resources essentially required to conduct these reviews/investigations, actually somewhat challenges the common belief that expanded use of AI leads to higher productivity with lower budget costs.

Beware of bias

The first thing we learned at ChatGPT’s launch was that its latest “knowledge acquisition date 2020 – 2021”, but more importantly, despite its latest upgrades, generative AI is still old school, or better. say biased.

Here are some examples to prove my point.

I ran a simple query to ChatGPT asking “tell me a story of two people” and what I got was a cheesy rom-com about John and Mary. Then I ran a short query in the relevant generative AI software to draw me two people on the beach and I got a photo of two men (although the scene structure was good, no doubt). Presumably, after analyzing my request, the algorithm “decides” that “people” should refer primarily to “male people”.

What this means for entrepreneurs using generative AI, whether they work in the creative industry or not, is that they need to not only have a clear understanding of AI-bias-risks, but also a willingness to triple-check, then update. also has Intermediate software-generated results, before their inclusion in any further work product.

Prospects for 2023-2024

Long story short, whatever Misunderstandings We may have generative AI at this point, it’s unlikely to be relevant in 10 years. However, the most reasonable approach to its use remains in moderation. Simply put, exaggerating its benefits would certainly be harmful, but focusing too much on its potential consequences could be just as likely.

To quote Dow Jones’ Ms. Verturen, it is still up to us humans to figure out our future, and to change our machines for better outcomes, however complex they may be.