{"id":203727,"date":"2025-10-05T18:26:06","date_gmt":"2025-10-05T18:26:06","guid":{"rendered":"https:\/\/www.newsbeep.com\/us\/203727\/"},"modified":"2025-10-05T18:26:06","modified_gmt":"2025-10-05T18:26:06","slug":"organoids-towards-energy-efficient-computing","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/us\/203727\/","title":{"rendered":"Organoids – Towards Energy Efficient Computing?"},"content":{"rendered":"

The BBC has reported on the advances made by scientists in Switzerland in the emerging field of Organoid Intelligence which uses human stem cells to build “mini-brains”, aka organiods.<\/p>\n


\n <\/p>\n

A major problem with the neural networks currently being used for artificial intelligence is their power consumption.\u00a0Training a large AI model on traditional hardware can use millions of watts of electricity. The human brain by contrast operates on just 20 watts.<\/p>\n

<\/p>\n

Starting with human skin cells, FinalSpark, a company based in Switzerland, is\u00a0growing tiny, lab-developed clusters of human brain cells referred to as “organoids”.<\/p>\n

<\/p>\n

\"mini-brains\"<\/p>\n

<\/p>\n

After months of cultivation, the organoids are connected to electrodes which\u00a0\u00a0allow electrical signals to be sent to and received effectively turning the cells into a component of a computing system.\u00a0They are prompted to respond to simple commands, with the resulting electrical activity tracked on a computer.<\/p>\n

<\/p>\n

The problem encounterd by such biological computers is keeping the organoids alive. Using a microfluidics system that supplies\u00a0 both water and nutrients, FinalSpark has managed to sustain them for up to around four months.<\/p>\n

<\/p>\n

The long-term goal is to train these organoids to learn and eventually perform complex tasks, leveraging their efficiency for next-generation AI systems.<\/p>\n

<\/p>\n

While FinalSpark is making an important contribution towards a biological brain capable of General Artificial Intelligence, it isn’t alone, nor is the endeavour new. Back in January 2022 we reported on DishBrain<\/a>, a biological neural network (BNN), that learned to play Pong in a simulated game world. Among the Related Articles in that report was\u00a0Google Has A Network More Like the Brain<\/a>, an item\u00a0from\u00a02019 which looked at Project Ihmehimmeli being undertaken by Google Research in Zurich.\u00a0<\/p>\n

<\/p>\n

Here’s a really deep question – if a organoid achieves general intelligence is it artificial general intelligence?<\/p>\n

<\/p>\n

\"mini-brainssq\"<\/p>\n

<\/p>\n

\u00a0<\/p>\n


\nMore Information <\/p>\n

Scientists grow mini human brains to power computers<\/a><\/p>\n


\nRelated Articles <\/p>\n

Brain In A Dish Learns To Play Pong<\/a><\/p>\n

<\/p>\n

Neurons Are Smarter Than We Thought<\/a><\/p>\n

<\/p>\n

Neurons Are Two-Layer Networks<\/a><\/p>\n

<\/p>\n

Google Has A Network More Like The Brain\u00a0<\/a><\/p>\n

<\/p>\n

To be informed about new articles on I\u00a0Programmer,\u00a0sign up for our\u00a0weekly newsletter<\/a>,\u00a0<\/a>subscribe to the\u00a0RSS\u00a0feed<\/a>\u00a0and\u00a0follow us on Twitter,<\/a>\u00a0Facebook<\/a> or\u00a0Linkedin<\/a>.<\/p>\n

\u00a0 <\/p>\n

\"Banner\"\/<\/a><\/p>\n


\n <\/p>\n

\"pico<\/a><\/p>\n


\n <\/p>\n

\u00a0<\/p>\n


\nComments <\/p>\n

Make a Comment or View Existing Comments Using Disqus <\/p>\n

<\/p>\n

or email your comment to: comments@i-programmer.info<\/a><\/p>\n

\u00a0