How The A.I. Industry Hurt Itself So Quickly

--

Photo by Gratisography

According to the tech industry hype machine that went into turbo overdrive over the past several months, designers, artists, lawyers, doctors and musicians should’ve been wiped out of the market. The existential angst of of dystopianism filtered into our living rooms, government hearings and boardrooms.

Some tech industry leaders proclaiming that they were going to cut, or had thousands of jobs due to A.I., chief among them the CEO of IBM. The constant deluge like a firehose in our faces from those proclaiming they’re ChatGPT and A.I. experts in just three months and can show you how to 200x your productivity are selling $300 PDF’s with lists of fragile tools that are at the whim of OpenAI API policies (it’s never good when 3rd party tools rely on a platform, Reddit being a prime case). None of these tools 200x anything but their bank accounts and drain yours.

Now, we hear rumours that ChaatGPT isn’t all it was cracked up to be. An insightful article by tech industry writer Alberto Romero takes you down that rabbit hole. It’s fascinating if true. It also supports just how overhyped and in the end, disappointing many of these A.I tools have been.

Sadly, this all has caused more hurt than benefit for those on the field of Artificial Intelligence. It is important to understand that there is no singular A.I. app at all. A.I. is an umbrella term for a toolbox with many different tools that do some very good things, used singularly or in various combinations.

Many of these tools have proven to be of great value to various industries, governments and in healthcare. Some A.I. tools, mixed the right way, can help us fight climate change and drastically improve our lifespans.

But OpenAI, perhaps Microsoft and Google too, in their schoolboy brawl to best each other for the sake of shareholder dividends and short-lived competitive advantage, while sagely proclaiming they’d fight the ethical dangers of A.I., yet dismissed their A.I. ethics teams have only ended up causing greater societal and cultural mistrust. They also created an artificial bubble of capital investments that are set to lose a lot of investors a lot of money.

As enterprise level businesses in Fortune 100 companies and smaller, come to realise these current A.I. tools just add to the current productivity paradox of Information Technologies, it hurts the A.I. companies that do have excellent produts that really can help. CIO’s and CTOs became ever more wary of the marketing and sales pitches A.I. companies make.

By their very nature CIOs and CTOs are skeptical of most any technology. They see through the marketing promises and when buying any platform technology, ERP software or Cloud service, largely ignore sales reps and seek out the truths told by sales engineers. Cutting through the marketing hype is their first action.

OpenAI, Midjourney and the rabid pack of snake oil scammers has perhaps, set back the real opportunities for A.I. for years. Perhaps no emerging, potentially revolutionary technology in modern history has been so quickly hyped and so quickly lost its lustre than ChatGPT and Generative A.I.

OpenAI and others now face multiple class-action lawsuits for copyright infringements. Their promises of being Good Shepherds ring hollow and are seen as simply a cynical ploy to safeguard their own market share and disrupt any potential competitors.

In some ways, this may also be a silver lining to a very dark cloud. Society and culture has become increasingly disillusioned and skeptical of many new technologies. This lesson, should A.I. toolmakers learn it, could help them to develop tools humans really want and deliver on the promises made.

A healthy dose of skepticism and critical thinking instead of the tech industry’s misaligned problem solving ideology of, ship fast and break things approach may have served their short-term interests. It is not serving their long-term interests now does it help their shareholder value in the end.

The disappointment of the launch and promise of Large Language Models and Generative A.I. has further eroded citizen and consumer trust in the technology industry and the technology giants. Apple, for example, clearly understands this as it has a much better understanding of human’s relationship with technology, and completely avoided even mentioning A.I. in their latest product announcements. The tech giants could learn from that.

For now, the lustre has worn off, the adrenaline, dopamine induced market and societal fears have drained away and found that the Emperor has no clothes.

--

--

Giles Crouch | Digital Anthropologist
Giles Crouch | Digital Anthropologist

Written by Giles Crouch | Digital Anthropologist

Digital Anthropologist | I'm in WIRED, Forbes, National Geographic etc. | Speaker | Writer | Cymru

No responses yet