In a groundbreaking development, researchers at Pacific Northwest National Laboratory (PNNL) have demonstrated how quantum-enhanced artificial intelligence (AI) can address some of the most pressing challenges facing humanity today. The collaboration has vigorously screened more than 32 million candidates for new battery materials. Their ultimate aim is to find safer, cheaper, and more environmentally friendly alternatives. This novel approach would even accelerate breakthroughs in material science and chemistry, possibly by years than earlier anticipated.
Quantum-enhanced AI leverages quantum-accurate data to enable faster, more accurate predictions. This executable model-driven approach is orders of magnitude more cost effective than classical computing. In addition, scientists are furiously working to use AI in tandem with quantum computing. Indeed, they think these tools will be fundamental to finding new materials and solutions to pressing problems such as climate change and disease.
The Importance of Quantum Computing in AI
One of the most exciting prospects of quantum computing in AI is its ability to completely change the game. This is particularly the case when electron correlation is an important effect. Electrons in strongly interacting systems form complex materials with exotic electronic properties. Approximating these systems requires advanced, nuanced simulations that classical computers are unable to rapidly produce.
For researchers, it’s clear that meaningful chemistry simulations will require hundreds, if not thousands, of high-quality qubits. She describes an ideal with these qubits having error rates closer to 10^-15 for the best results. Reaching this threshold level of reliability requires resulting fault tolerance with redundant encoding of quantum information into logical qubits. As a reminder, each logical qubit can consist of hundreds of physical qubits. To ensure precision, we really require on the order of one million physical qubits.
As John P. Perdew wrote back in the summer of 2001, an AI revolution in scientific research has been the dream of many for the last 60 years. Recent breakthroughs demonstrate a huge step in the right direction within this thrilling field.
“Set up on the earth, and the top of it reached to heaven. And behold the angels of God ascending and descending on it.” – John P. Perdew
Accelerating Material Discovery
The PNNL research team has significantly reduced that search from 32 million possible battery materials. In under a week, they’ve already nailed down 800 of these candidates as highly promising. Conventional approaches would have required about two decades to test such a broad range of possible candidates. With quantum-enhanced AI, researchers are better able to identify “first-time right” candidates. This strategy steers just the most promising molecules down the path of laboratory synthesis and testing.
Along with saving time, this rapid pace leads to severe cost savings in the process of discovering material. Quantum-enhanced AI flattens the simulation curve by allowing extremely fast predictions for related systems. This ambitious achievement deepens the boundaries of what’s achievable in scientific discovery.
Large language models help researchers rapidly sift through massive sets of data. This allows them to spend their time working on the materials that are most likely to be a success. This calculated and targeted approach has the potential to produce the radical advances in battery technology that are necessary for all sustainable energy solutions.
Implications for the Future of Scientific Research
When quantum computing is paired with AI, the two technologies can transform scientific research. This blending will change every discipline profoundly. As a result of advancements in computational power, researchers are able to perform higher fidelity simulations and model more complex interactions than ever before. This paradigm shift would make transformative breakthroughs possible across disciplines, from materials science to biomedical research.
Quantum computing is getting more reliable and efficient by the day. As this technology evolves, its pairing with AI will certainly change how scientists approach scientific problem-solving. The ramifications go far beyond just finding new materials—their work could impact faster drug development, better climate modeling, and more.