Trained as a physicist and biophysicist, Anthropic CEO Dario Amodei once set his sights on curing disease. In fact, building large language models was never part of the scheme. Then, Biology, the field he hoped would unlock medical breakthroughs, ultimately pushed him towards artificial intelligence.
“The thing I noticed about studying biology was its incredible complexity,” Amodei told Zerodha co-founder Nikhil Kamath on his podcast 'WTF Is'.
Working on protein mass spectrometry to identify protein biomarkers, he confronted a system that seemed endlessly layered. “If you look at the protein mass spec work that I did, trying to find protein biomarkers. It's just really incredible how much complexity there is,” he told Kamath.
Proteins are not static molecules; they change form and function in intricate ways.
“You have a given protein, it's like the RNA gets spliced in a whole bunch of different ways depending on where it is in the cell. Then it gets post-translationally modified, phosphorylated, complexed with a whole bunch of other proteins. And I was starting to despair that it was too complicated for humans to understand," he said.
Around 15 years ago, something else caught his attention: early breakthroughs in deep learning. In particular, AlexNet, one of the first neural networks to outperform traditional methods in image recognition, was the start of AI incorporation.
AI systems, he realised, shared some characteristics with the human brain. But unlike biological brains, they could scale in size, data exposure, and learning capacity. If neural networks could learn to recognise images, perhaps they could also learn the language of proteins, genes, and cellular systems.
“Maybe this is ultimately going to be the solution,” Amodei said.
He went on to work with Andrew Ng at Baidu, then spent a year at Google. Shortly after OpenAI was founded, he joined and eventually led its research efforts for several years.
On the podcast, he also described the current AI moment as standing on a shoreline, watching a massive wave gather in the distance. “It's as if this tsunami is coming at us,” he said. “It's so close, we can see it on the horizon, and yet people are coming up with explanations like, ‘Oh, it's not actually a tsunami, it's just a trick of the light.'”
In his view, AI systems such as Anthropic's Claude are rapidly approaching human-level performance in many cognitive tasks.
