The lure of (de)generative AI

Korean-German philosopher Byung-Chul Han opines that we have gone from being homo-sapiens to ‘phono-sapiens’.
Last Updated 24 February 2024, 22:08 IST

In the course of my readings, I came across certain memorable phrases and sentences that perfectly describe the evolution of today’s digital society from its analogue version of some 30 years ago. In the words of French philosopher Paul Virilio, “When you invent the ship, you also invent the shipwreck; when you invent the plane, you also invent the plane crash; and when you invent electricity, you invent electrocution.” He also remarked, “Digital messages and images matter less than their instantaneous delivery; the shock effect always wins out over the consideration of the informational content.”

Korean-German philosopher Byung-Chul Han opines that we have gone from being homo-sapiens to ‘phono-sapiens’, a perspective that succinctly captures our current obsession with smartphones. In Han’s view, “power operates more effectively when it delegates surveillance to discrete individuals”.

The ever-increasing need to have a smartphone to navigate daily life has been destroying our personal and professional lives without our being aware of it. As one astute Guardian columnist observed, “The smartphone is Catholicism with better technology, a modern rosary that is handheld confessional and effective surveillance apparatus in one.” Even those of us who are aware have done nothing to counter this since we are at the mercy of the overlords of the IT industry. Gone are days when our DNA, our genetic code used to define who we are as individuals, having been supplanted by QR codes and credit scores. Now that generative AI, which can instantly create text, video, images and sounds, is upon us and the IT industry expects us to embrace it, we can expect to be submerged in deep fakes and misinformation as never before and, worse yet, pretend it isn’t happening.

If you were to look up the meaning of the word ‘GAIA’ in the Merriam-Webster Dictionary, you will be surprised to learn that it is not an acronym for “Generative AI for All’ but the hypothesis that the living and non-living components of earth function as a single system in such a way that the living component regulates and maintains conditions (such as the temperature of the ocean or composition of the atmosphere) so as to be suitable for life. This symbiosis is simply absent from the artificial intelligence versus human intelligence debate.

While generative AI, as it currently exists, stands apart from search engines and databases of whatever kind, it can easily produce web pages and multimedia databases of dubious content which can then be indexed by Google, Bing, Meta and other search engines and show up as legitimate items in search results in response to user queries. Google prides itself on being able to rate websites based on four criteria -- experience, expertise, authoritativeness and trustworthiness. Since Google accounts for 89% of all web searches, it is disconcerting to note that Google has become the ultimate arbiter of which websites to rely on and which ones to avoid. I seriously doubt Google applies the same criteria when it accepts advertising from some companies but not others.

Scores of companies across the globe have been creating new Large Language Model (LLM)-based products similar to ChatGPT. For example, Core42, an Emirati company, has created JAIS-30b by collecting Arabic data from multiple sources, including web pages, Wikipedia articles, news articles, Arabic books, and social network content. According to the company, this data has been augmented by data generated from translating “high-quality English resources such as English Wikipedia and English books” using an in-house machine translation system. The company’s majority shareholder is the UAE’s National Security Adviser!

In a recent, self-serving op-ed in the New York Times, Jim Albrecht, senior director of news ecosystem products at Google from 2017 to 2023, opined that new technology has created a landscape where AI outfits might not need to link to news sites at all. Instead, they could just take the news, have a robot rewrite it and publish it in their own products. He also wrote that news media’s most important work was the discovery and verification of new information and not forcing the tech industry to make blanket payments for rewrites of what is already long known.

Like the gun industry, IT companies have enjoyed almost total immunity from civil and criminal prosecution even though their products wreaked havoc on society at large. And they get to keep their enormous profits. Since the mandate of Competition Commissions is to prevent the creation of monopolies and oligopolies, the products themselves, and the potential harm caused by them, never come under the scanner.

As for India’s IT mimic men, they will soon be at it again, only this time in (de)generative AI.

(Roger Marshall is a computer scientist, a newly minted Luddite and a cynic)

(Published 24 February 2024, 22:08 IST)

Follow us on