Scientists Are Frighteningly Close to Achieving Synthetic Biological Intelligence
Tim Newcomb
Mon, July 24, 2023
‘Synthetic Biological Intelligence’ is Close
Yulia Reznikov - Getty Images
A group of researchers have already grown brain cells on silicon chips and then taught them to perform tasks.
This merging of artificial intelligence and synthetic biology opens a new realm of continual machine learning.
A $600,000 AUD Australian national intelligence grant continues to the research into the “cyborg computing chips.”
Lab-grown synthetic brain cells can already learn tasks. Now, the same team that brought us 800,000 Pong-playing brain cells living in a dish has received $600,000 AUD from Australia’s National Intelligence and Security Discovery Research Grants Program to further push these lab-grown brain cells embedded onto silicon chips into the world of machine learning.
This entire project “merges the fields of artificial intelligence and synthetic biology to create programmable biological computing platforms,” Adeel Razi, associate professor at Monash University’s Turner Institute for Brain and Mental Health, says in a news release. “This new technology capability in future may eventually surpass the performance of existing, purely silicon-based hardware.”
The official language from Australian government’s Office of National Intelligence when listing the three-year research project spells it out: “The project aims to merge the fields of artificial intelligence and synthetic biology to create programmable cyborg computing chips.”
How’s that for a future?
As part of that previous Neuron article on the Pong-playing capabilities, the research team said that synthetic biological intelligence “previously confined to the realm of science fiction” is close.
While AI capabilities are one thing, Razi’s research looks to take machine learning to the next level. Real human brains are known for their ability to handle lifelong learning, while current AI systems suffer from “catastrophic forgetting,” Razi says, when adding new tasks to their repertoire. The continual lifelong learning abilities of these synthetic chip brains allow machines to acquire new skills without comprising old ones. The chips can also adapt to changes and apply previously learned knowledge to new tasks, all while conserving computing power, memory, and energy.
“The outcomes of such research would have significant implications across multiple fields such as, but not limited to, planning, robotics, advanced automation, brain-machine interfaces, and drug discovery,” Razzi says, “giving Australia a significant strategic advantage.”
The Monash University group, led by Razzi, has partnered with Cortical Labs of Melbourne to continue the research. Now, they hope to grow human brain cells in a lab dish called the DishBrain system to better understand the biological mechanisms that underpin this continual learning ability.
“We will be using this grant to develop better AI machines that replicate the learning capacity of these biological neural networks,” Razzi says. “This will help us scale up the hardware and methods capacity to the point where they become a viable replacement for in silico computing.”
A group of researchers have already grown brain cells on silicon chips and then taught them to perform tasks.
This merging of artificial intelligence and synthetic biology opens a new realm of continual machine learning.
A $600,000 AUD Australian national intelligence grant continues to the research into the “cyborg computing chips.”
Lab-grown synthetic brain cells can already learn tasks. Now, the same team that brought us 800,000 Pong-playing brain cells living in a dish has received $600,000 AUD from Australia’s National Intelligence and Security Discovery Research Grants Program to further push these lab-grown brain cells embedded onto silicon chips into the world of machine learning.
This entire project “merges the fields of artificial intelligence and synthetic biology to create programmable biological computing platforms,” Adeel Razi, associate professor at Monash University’s Turner Institute for Brain and Mental Health, says in a news release. “This new technology capability in future may eventually surpass the performance of existing, purely silicon-based hardware.”
The official language from Australian government’s Office of National Intelligence when listing the three-year research project spells it out: “The project aims to merge the fields of artificial intelligence and synthetic biology to create programmable cyborg computing chips.”
How’s that for a future?
As part of that previous Neuron article on the Pong-playing capabilities, the research team said that synthetic biological intelligence “previously confined to the realm of science fiction” is close.
While AI capabilities are one thing, Razi’s research looks to take machine learning to the next level. Real human brains are known for their ability to handle lifelong learning, while current AI systems suffer from “catastrophic forgetting,” Razi says, when adding new tasks to their repertoire. The continual lifelong learning abilities of these synthetic chip brains allow machines to acquire new skills without comprising old ones. The chips can also adapt to changes and apply previously learned knowledge to new tasks, all while conserving computing power, memory, and energy.
“The outcomes of such research would have significant implications across multiple fields such as, but not limited to, planning, robotics, advanced automation, brain-machine interfaces, and drug discovery,” Razzi says, “giving Australia a significant strategic advantage.”
The Monash University group, led by Razzi, has partnered with Cortical Labs of Melbourne to continue the research. Now, they hope to grow human brain cells in a lab dish called the DishBrain system to better understand the biological mechanisms that underpin this continual learning ability.
“We will be using this grant to develop better AI machines that replicate the learning capacity of these biological neural networks,” Razzi says. “This will help us scale up the hardware and methods capacity to the point where they become a viable replacement for in silico computing.”
No comments:
Post a Comment