A team of 16 experts worked nonstop to replicate the quantum results manually.
HSBC has claimed a global first in financial tech after using IBM’s quantum chip to increase bond price prediction accuracy by 34%, according to information provided by the bank on Thursday.
The trial, done with IBM’s Heron quantum processor, ran real, anonymized European bond trading data through the system, not fake models or simulations, and resulted in a major accuracy jump. That’s the first time any bank has used quantum to process real trading data at full scale.
The test didn’t involve live trades, but it wasn’t theoretical. HSBC confirmed it was a production-scale demonstration. That means this wasn’t just another one of those tech tests that sit in a sandbox forever. This was finance in the wild, just without actual money moving. The bank believes this step sets the tone for how quantum will start creeping into daily financial systems worldwide.
The quantum test focused on over-the-counter bond markets, where trades happen privately between two firms, no exchange involved. These are harder to predict due to low transparency and fragmented data. That’s exactly why HSBC picked this use case. IBM’s Heron processor, the latest in its quantum lineup, handled the task by running calculations in parallel, unlike classical chips that go line by line.
Philip Intallura, HSBC’s head of quantum tech, said, “Is this a ‘Sputnik moment’ for quantum? My instinct is yes.” He pointed to how this milestone could trigger a wave of fast-moving adoption. “It will create a flurry of activity,” Philip added, hinting that rivals will rush to close the gap.
Josh Freeland, HSBC’s global head of algo credit trading, gave a glimpse into what was happening behind the scenes. He said the trial involved 16 experts (physicists, machine learning engineers, AI specialists) who were “working around the clock” to replicate what the chip had done. “If you could get something like this result every day, that would be quite something,” Josh said. He explained that in trading, even a single-digit gain, when repeated thousands of times, can move the needle in a big way.
This tech push isn’t limited to HSBC. Other big names like JPMorgan, Goldman Sachs, and Citigroup are also pouring resources into quantum. In March, JPMorgan said it created and validated truly random numbers with a quantum computer built by Quantinuum. The bank claims this tech will help improve encryption, security, and maybe even trading systems. That work was later published in Nature, a top science journal.
Meanwhile, Google’s Willow quantum chip, separate from the HSBC effort, solved a specific mathematical task in five minutes that classical supercomputers couldn’t solve in the entire age of the universe. That kind of speed is what has the financial sector throwing billions into quantum development, even if widespread rollout still seems distant.
Quantum revenue is expected to skyrocket to $72 billion in the next ten years, up from just $4 billion last year, according to consulting firm McKinsey. The report, published in June, lists finance alongside industries like life sciences and chemicals as key drivers. Henning Soller, who leads McKinsey’s quantum research in Frankfurt, said the tech’s value is clearest when it’s applied to pricing predictions. In finance, every percentage counts.
“If one bank is able to start using quantum computing to develop a program, then the others will be developing it the next day and people will not sleep until they have it,” said Miklos Dietz, McKinsey’s senior partner in Vancouver. He didn’t hold back on what’s coming next. “When it comes, it will be explosive.”
HSBC isn’t pretending the tech is perfect yet. But it believes it’s on the edge of something real. Philip said the work is proof that banks don’t have to wait five or ten years to see results. “We have great confidence we are on the cusp of a new frontier of computing in financial services, rather than something that is far away in the future.”
Don’t just read crypto news. Understand it. Subscribe to our newsletter. It’s free.
