The Giants of AI: Google, OpenAI, Microsoft, and Meta
by Tommaso Saso
If artificial intelligence were a game of poker, there woudl be few players at the main table, and everyone would have piles of astronomical chips! This is because developing the cutting edge requires rare talent combinations, data, calculation power and, above all, money – a lot of money.Welcome to the world of the AI giants, where the future of humanity is decided in Silicon Valley meeting rooms.
Google is the noble grandfather of this exclusive club. Its domain began almost by chance. Two students,Larry Page and Sergey Brin,simply wanted to create a better search engine for their doctoral thesis at Stanford. Unknowingly, they were creating the most elegant artificial intelligence system in the world. Each Google search is a dialogue wiht an AI that indexes billions of web pages and anticipates what you really want to know.
But Google didn’t stop at search. In 2014, it purchased DeepMind for $500 million, bringing on board the minds that created alphago. It then developed TensorFlow, the most used machine learning framework globally, and built TPUs – chips designed specifically for AI. Google also created Google Translate, which instantly translates between 108 languages and is used daily by many.
Then came OpenAI. Founded in 2015 by a group of visionaries led by Sam Altman and Elon Musk, it was born with a noble mission: to develop safe AI for the benefit of all humanity. “open” signified openness, transparency, and research sharing.It was David against Goliath – an idealistic startup against corporate giants.
The project became reality in November 2022, when OpenAI released ChatGPT. It reached one million users in five days and one hundred million in two months! No technological product in history has ever experienced such explosive growth. Suddenly, everyone was talking about AI. ChatGPT became the new Google, the reference point for revolutionary technology.
But success has a price, and OpenAI discovered that idealism is costly. Developing GPT-4 cost over $100 million, and the infrastructure to run ChatGPT costs millions per day. The “Open” company became increasingly “Closed,” forced to forge billion-dollar partnerships with Microsoft. Elon Musk subsequently left, accusing OpenAI of betraying its original mission by becoming a commercial enterprise.
Meanwhile, Microsoft, to compete directly, invested $13 billion in OpenAI, gaining exclusive access to its technologies and integrating ChatGPT into its Bing search engine, challenging Google’s monopoly in search. Furthermore, with Copilot, it integrated AI into Office applications like Word and Excel, transforming everyday tools into clever assistants. Economists call this the strategy of the “fast follower” – letting others innovate, then acquiring that innovation.
Meta (formerly Facebook) has instead chosen a different approach: strategic open source. In 2023, it released Llama, a powerful language model comparable to GPT-4, but completely free. This may seem generous,but it’s a refined strategy! If everyone uses your tools,you control the standards.It’s like giving away the coffee machine and then selling the pods. Meta recognizes it cannot currently match the resources of giants like Google and OpenAI, but it can attempt to democratize AI and become indispensable.
Mark Zuckerberg has also made a bold bet on the metaverse, investing over $100 billion in virtual reality and AI. So far, the results are…