Mark Thomson, the future general manager of CERN (European Nuclear Research Organization), says artificial intelligence opens the way of huge progress in particle physics.
Photo artificial intelligence: shutterstock
Advanced artificial intelligence will revolutionize fundamental physics and could open a window to the fate of the universe, according to the future general manager of CERN (European Organization for Nuclear Research).
Professor Mark Thomson, the British physicist who will take over the CERN leadership on January 1, 2026, states that automatic learning opens the path of progress in the particle physics that promises to be comparable to the prediction of protein-based protein structures, which has brought Google Deepmind the prize Nobel in October.
At Large Hadron Collider (LHC), he said, similar strategies are used to detect incredibly rare events that hold the key to how the particles have acquired meals in the first moments after the Big Bang and if our universe could be on the verge of a he collapsed catastrophically.
“These are not incremental improvements,” Thomson said. “These are very, very, very large improvements that people make by adopting very advanced techniques. ”
“It will be quite transformer for our field,” he added. “They are complex data, like folding proteins – which is an incredibly complex problem – so if you use an incredibly complex technique, such as you, you will win.”
The intervention comes in the context in which the CERN Council presents arguments in favor of the future circular accelerator, which, at a circumference of 90 km, would eclipse the LHC. Some are skeptical, given the lack of extraordinary LHC results since the historical discovery of the Higgs Boson in 2012, and Germany described the proposal of $ 17 billion as uninhabil. Thomson said, however, that artificial intelligence has given a new impulse to seeking new physics on a subatomic scale and that major discoveries could take place after 2030, when a major update will increase the intensity of the LHC beam ten times.
This will allow unprecedented observations of the Higgs boson, nicknamed the particle of God, which gives mass to other particles and connects the universe between them.
“There is a particular measure about the Higgs boson that is so fundamental to the nature of the universe ”, Thomson said. “What we will follow is the production of not a Higgs Boson, but two Higgs bosons. “
This, he said, will allow scientists to measure for the first time the way the Higgs particle gives its own mass – a phenomenon called Higgs self -council.
Two Higgs Bosons appear simultaneously so rarely, and the particles are so evasive – disintegrating into more familiar particles – that, five years ago, Thomson said it would have supposed that this exceeds the LHC capabilities. “Now I am confident that we will make a good measurement ”, Thomson said.
The power of Higgs self -council is crucial to understand how, at a second trillion after the Big Bang, a change in the Higgs field has led to the sudden acquisition of mass by particles. It could also reveal whether the Higgs field has reached a final, stable, resting state or if another drastic transition could take place in the future, a scenario that would lead to almost instant evaporation of the universe as we know it. The standard model of physics suggests that this is a possibility – but there is no need to alarm.
“It is not something that could happen on a scale of time that has any relevance even for our stars”, said Dr. Matthew McCullough, theoretician physicist at CERN. “So it has no connection with humanity in this regard. On the other hand, is it a scientific question – could this happen? ”
According to Thomson: “It is a very deep fundamental property of the universe, which we do not fully understand. If we see that the Higgs self -council is different from our current theory, this would be another massive, massive discovery. And you don’t know until you do the measure. “
Artificial intelligence is injected into every aspect of LHC operation, from the decision to be collected to how they need to be interpreted. “When LHC collides protons, it produces about 40 million collisions per second and we have to make a decision within a microsecond … which events are interesting and that we want to keep and to throw them. “said Dr. Katharine Liey, who works at the LHC Atlas experiment. “With the data we collected we do more than we thought we could do with 20 times more data ten years ago. So, I advanced at least 20 years. Much of this is due to artificial intelligence. ”
Scientists have long hoped that LHC could be able to produce dark matter, a substance that is believed to be a large part of the universe. But given that the nature of dark matter is completely unknown, its search is a difficult task. According to Thomson, the generative artificial intelligence could help to unravel this puzzle. “You can start asking more complex, open questions,” he said. “Instead of looking for a certain signature, do you ask the question: Is there something unexpected in this data?”
