AI Tokens Outperform CoinDesk 20 Index Tuesday Despite Clouds Forming Over Industry
New research from Goldman Sachs and Sequoia challenges assumptions that Artificial Intelligence and Large Language Models can change the world.
- AI tokens outperformed the broader market early Tuesday.
- This comes despite clouds forming over the industry and growing talk of a bubble.
Artificial Intelligence (AI) tokens outperformed the CoinDesk 20 Tuesday, a broad measure of the performance of the largest digital assets on the market, despite bearish reports from major financial firms that argue AI's growth potential is constrained.
CoinGecko data shows that the AI token category was up 4.5% Tuesday, while the CoinDesk 20 was up just under 1.5%.
Major AI tokens include NEAR (NEAR), Internet Computer (ICP), Render (RNDR} and SingularityNet AGIX).
This market move comes after Goldman Sachs released a report that criticized generative AI for limited productivity benefits, uncertain returns, and high power demands, questioning its economic viability.
In particular, Goldman's research highlights the massive jump in power demand that AI will require. They estimate that by 2030, power demand from data centers will jump from 3% to 8% of the total share of the grid.
"The US has unfortunately lost the ability to build large infrastructure projects—this is a task better suited for 1930s America, not 2030s America," the report quotes Brian Janous, a former Vice President of Energy at Microsoft, as saying. "So, that leaves me a bit pessimistic," he continued.
Janus argues that utilities and policymakers are investing in transmission infrastructure to adapt to new energy sources, but key to their ability to pull this off is a massive overhaul of the grid while keeping consumer rates flat – or the politics around the issue will sour.
"On balance, I’m optimistic that America can rise to meet the challenge, though the next decade will likely prove painful as the demand for power outpaces the available supply," he concluded.
Meanwhile, another report, this time from Sequoia argues that there's still a significant revenue gap in the sector, coming in at $600 billion.
The report's author emphasizes that AI will create a "huge amount of economic value," but first, we need to tread through the speculative frenzy.
"Those who remain level-headed through this moment have the chance to build extremely important companies," the author, David Cahn, writes. "But we need to make sure not to believe in the delusion that has now spread from Silicon Valley to the rest of the country, and indeed the world."
AI tokens certainly have the potential to be some of these extremely important companies. Nvidia has recognized this, inviting NEAR's founder Illia Polosukhin on-stage at its GTC Conference earlier this year for a fireside chat with Jensen Huang. NEAR revolutionized natural language processing with its research on transformers, the underlying architecture behind LLMs.
But crypto has also been known for its speculative frenzy. It remains to be seen if AI can build around it.
Sam Reynolds
Sam Reynolds is a senior reporter based in Taipei. Sam was part of the CoinDesk team that won the 2023 Gerald Loeb award in the breaking news category for coverage of FTX's collapse. Prior to CoinDesk, he was a reporter with Blockworks and a semiconductor analyst with IDC.