A magnified image of a brain organoid produced in Thomas Hartung’s lab, dyed to show neurons in magenta, cell nuclei in blue, and other supporting cells in red and green.
Jesse Plotkin/Johns Hopkins University
Artificial intelligence seems to be, but a team of scientists argues that something called “organoid intelligence,” or OI, powered by living human brain cells could one day outperform any artificial system, and do it far more efficiently too.
Organoids are three-dimensional clumps of biological tissue that scientists have been growing and experimenting with for years. Researchers led by environmental health sciences professor Thomas Hartung at Johns Hopkins University are working with brain organoids that could lead to the development of a “biocomputer” powered by human brain cells.
“We are at a moment in time, where the technologies to achieve actual biocomputing have matured,” Hartung told me via email. “The hope is that some of the remarkable functionalities of the human brain can be realized as OI, such as its ability to take fast decisions based on incomplete and contradictive information (intuitive thinking).”
Hartung and colleagues unveil an extensive vision for the future of OI on Tuesday in the journal Frontiers in Science.
The team includes scientists from Cortical Labs, which made headlines last year for creating a dish full of live brain cells that quickly.
Using organoids grown from cells is advantageous for scientists because it doesn’t require human or animal testing. Hartung has been creating functional brain organoids since 2012 using human skin cells that are reprogrammed into an embryonic stem-cell like state. They can then be used to form brain cells and, eventually, organoids with functioning neurons and other features that can sustain basic functions like memory and continuous learning.
“This opens up research on how the human brain works,” Hartung said in a statement. “Because you can start manipulating the system, doing things you cannot ethically do with human brains.”
A living computer
He and his colleagues envision assembling brain organoids into new forms of biological computing hardware far more energy efficient than current supercomputers.
“The brain is still unmatched by modern computers,” Hartung said. “Frontier, the latest supercomputer in Kentucky, is a $600 million, 6,800-square-feet installation. Only in June of last year, it exceeded for the first time the computational capacity of a single human brain — but using a million times more energy.”
Hartung concedes that computers are faster at processing numbers and data but maintains that brains remain better when it comes to complex logical problems.
“Computers and the brain are not the same, even though we tried making computers more brain-like from the beginning of the computer age. The promise of OI is to add some new qualities.”
Concepts like biological computers and organoid intelligence could lead to a library’s worth of new ethical discussions. Conversations about organoids becoming sentient, conscious or self-aware and the ensuing implications have been underway for years now, even though the technology is thought to be immature at the moment.
“There is probably no technology without unintended consequences,” Hartung told me. “While it is difficult to exclude such risks, as long as humans control input and output as well as the feedback to the brain on the consequences of its output, humans have control. However, like AI, the problem comes as soon as we give AI/OI autonomy. Machines, whether based on siliceous or cellular machinery, must not decide about human life.”
Members of the research team with backgrounds in bioethics have been working to assess the ethical implications of working with OI.
Organoid intelligence and biocomputers won’t pose a threat to AI or human brains grown the old-fashioned way anytime soon. But Hartung believes it’s time to begin increasing production of brain organoids and training them with AI in order to breakthrough some of the shortcomings of our existing silicon systems.
“It will take decades before we achieve the goal of something comparable to any type of computer,” Hartung said. “But if we don’t start creating funding programs for this, it will be much more difficult.”