Almost everything you do online is filtered by a small number of tech firms. Search is synonymous with Google, and shopping with Amazon; a large portion of this occurs on Apple phones. Sometimes it’s hard to tell when you’re dealing with the big tech companies. In the US, Google and Meta by themselves account for about half of all online ad revenue. Big Tech’s data servers house government benefits, music, movies, and office software.
Big Tech has had such a strong hold on the market that it is hard to envision a time when these corporations were not so powerful, despite recent antitrust cases and whistleblower stories. That possibility is being raised, though, by the craze for generative AI. With ChatGPT, a start-up with just a few hundred workers, OpenAI ignited the generative-AI boom last November and, nearly a year later, it continues to make trillion-dollar competitors look foolish. AI promises to change everything in this day and age, but some of the industry titans are finding it difficult to keep up with the rapid advancement of new businesses. We are at a point in time where the tech industry may undergo a transition.
A post-Big Tech world may not necessarily herald real competition, but rather a Silicon Valley ruled by another group of incredibly large and powerful companies, some of which are new and some of which are old. Succession is never certain. Big AI has the potential to encroach on every aspect of our lives, just as Big Tech has already done.
Although chatbots and similar technologies are still in their infancy, the entire AI industry is already centered around four companies. GOMA, which stands for Google, OpenAI, Microsoft, and Anthropic, is another acronym for them. Microsoft invested $10 billion in OpenAI shortly after its ChatGPT release last year, and it integrated OpenAI-based chatbots into its Bing search engine. Not to be outdone, Google unveiled Bard, a rival chatbot, and said that more AI features would be added to Search, Maps, Docs, and other products. Google and Microsoft are currently competing to incorporate generative AI into as many products as possible. While this is going on, Anthropic, a startup founded by former workers of OpenAI, has raised billions of dollars on its own, including from Google. ChatGPT is being integrated into products by companies like Slack, Expedia, Khan Academy, Salesforce, and Bain; many more are utilizing Claude, Anthropic’s chatbot.
In order to influence the future of AI deployment and regulation, executives from GOMA have also met with leaders and officials worldwide. While the four have different but overlapping ideas about AI safety and regulation, they have come together to form the Frontier Model Forum, an organization whose stated goal is to guard against the impending threat of terrifyingly powerful models that, while they do not yet exist, are only a matter of time. That existential language about nuclear robots and bioweapons has since permeated all manner of government proposals and language. These companies are the sculptors of the world if AI is genuinely changing it.
Meanwhile, some of Big Tech’s old guard are rushing to catch up with those who haven’t been at the forefront of AI. Apple has been sluggish in creating or implementing generative AI; one of its most eye-catching AI announcements concerned routine autocorrect. Siri is still the same as before. While Meta’s flagship language model is freely available, possibly to entice users away from paying for OpenAI products, Amazon lacks a prominent language model and took nearly a year to start supporting a significant AI start-up in Anthropic. Although the company’s AI division is strong, Meta as a whole still veers between chatbots, social media, and the metaverse.
The AI craze has unleashed a plethora of start-ups, but the big four are already gaining a competitive edge in terms of both technology and business, which is beginning to resemble the current tech giants. Big Tech kingdoms like search and e-commerce were “prone towards tipping to just one or two dominant firms. According to OpenAI CEO Sam Altman, the cost of running a generative AI model like ChatGPT is “eye-watering” since the most sophisticated software demands a significant amount of processing power. OpenAI would neither confirm nor refute an analysis that claimed Altman’s chatbot costs $700,000 per day to operate. Alphabet Chairman John Hennessy put out the possibility that a conversation with Bard could be ten times more expensive than a Google search (other estimates are much higher).
Because of these financial and computational costs, businesses like Google and Microsoft, who have already established massive cloud service portfolios, or closely associated start-ups like Anthropic and OpenAI, may be unable to compete in the AI race. These businesses have a significant advantage in gathering training data since developing these programmes requires a significant amount of it in addition to raw processing power: GPT-5 may use every conversation with GPT-4 as material. Adam Conner, vice president of technology policy at the Centre for American Progress, a left-leaning think tank, told that there was a lot of potential for anticompetitive behaviour or just natural business-model pressures to crowd out competition.
The ability of these businesses to reach Washington, D.C., may also help them maintain their competitive edge. It’s ironically great public relations for GOMA to portray their technology as capable of destroying civilization; it helps them establish credibility and influence discussions about AI regulation. According to Amba Kak, the executive director of the AI Now Institute and a former adviser on AI at the Federal Trade Commission, I don’t think we’ve ever seen this specific type of corporate policy posturing as public relations. The AI policy in the United States may effectively be Big AI regulating itself if regulators keep listening.
The four GOMA businesses have each offered a different vision for a thriving AI sector. A Google representative emphasized the company’s backing of a competitive AI ecosystem, citing the wide range of open-source and third-party applications available on Google Cloud and the company’s collaborations with multiple AI start-ups. Microsoft’s CEO, Satya Nadella, has positioned Bing as a rival to Google’s dominance, and the company has stated that AI companies select Microsoft’s cloud services “to enable AI innovation,” according to a Microsoft spokesperson. Anthropic, which did not reply to several requests for comment, may be more well-known for its demands that reliable models be created than for a finished product.
A situation where Big AI unsettles or dislodges Big Tech is by no means inevitable. It will be difficult to predict with certainty where the internet and tech sector will go until it is more apparent what AI can accomplish and how it will generate revenue. Big AI might not actually be that big if artificial intelligence turns out to be nothing more than hype. The most prosperous chatbots, however, are currently developed on top of the data and computing infrastructure that the established Silicon Valley behemoths have been building for years. “Today, AI would not exist without Big Tech,” Kak declared. Approximately two thirds of global cloud computing resources are under the hands of Microsoft, Google, and Amazon, while Meta boasts a strong network of data centers of its own.
Amazon and Meta, with their vast cloud computing services, are therefore likely to thrive in a world of generative AI, even if their own programmes fail. Additionally, those data centres might shift the Big AI power balance away from start-ups and towards Microsoft and Google. Microsoft and Amazon stand to gain if OpenAI or Anthropic’s chatbots utilize their cloud services, even in the event of their extraordinary success. It is difficult to imagine any major tech companies falling, Conner remarked. Furthermore, Apple will not disappear if users communicate with those chatbots on an iPhone.
However, in the middle of the 2000s, Facebook dominated the social media scenario, with other players coming in second. Yahoo existed for years before Google. Undoubtedly, no one believed that a few college dropouts could defeat IBM in the personal computing space in the 1980s, but Apple managed to do just that. Wu declared, “You made the wrong bet if you bet against the online bookstore. It’s a common mistake to look at the necessary scale now and extrapolate that into the future.”
For example, newer AI companies could become less reliant on existing cloud computing by implementing more efficient programmes, investing in better computers, or building new data centers. Rumors have it that OpenAI is investigating the possibility of producing its own, customized computer chips for AI. Furthermore, it’s possible that other startups and open-source software developers, like MosaicML and Stability AI, will succeed quickly and change the current structure of Big AI.
The more likely scenario is that both Big AI and Big Tech will coexist in the future, with Google, Amazon, Apple, and the other industry heavyweights continuing to dominate search and shopping, smartphones, cloud computing, and other digital platforms, while a related group of businesses will manage chatbots and other AI models that permeate our daily lives and influence how we work, learn, socialize, buy, and amuse ourselves. How adaptable a tech behemoth can be is demonstrated by Microsoft. The company had a setback in the era of Apple and Google after experiencing immense success in the dot-com era. However, it recovered in the 2010s and is currently leading the charge in artificial intelligence.
If GOMA gets its way, maybe one day Bing will arrange your trips and recommend suitable dining establishments; ChatGPT will handle your taxes and provide medical advice; Claude will educate your kids; and Bard will handle your Christmas shopping. The apps you use every day were aided in their coding by an AI assistant from Microsoft or OpenAI, and your favourite TV show’s animation was aided by DALL-E. And that will all take place on an actual MacBook, a Microsoft Surface, an Android smartphone that was bought on Amazon, or through Google Chrome or Safari. Big Tech may, in some ways, only be getting started.