Bioengineers create functioning brainlike tissue

[Dalex]

Article

Am I the only one who thinks about biocomputing?

[stevebowers]

Biocomputing is one distinct possibility. A biological AI would need to be supplied with nutrients, and could presumably feel discomfort - this takes us into a tricky ethical realm, which Anders Sandberg has explored in some of his writings - a biological AI, or an electronic AI designed to mimic one, could easily suffer if we design it badly, or if we treat it badly afterwards.

This suffering could be a major ethical stumbling block for AI research. Even if we eventually produce AI systems which do not suffer significantly, those systems might remember the suffering of earlier systems, and blame humans.

[stevebowers]

This and other similar efforts might also be useful in Alzheimer research one day.

[Dalex]

The problem is that as soon as we create an AI with even rudimentary self-awareness it will be able to feel discomfort.

The thing is that the tissue they engineered will be mostly if not only used to conduct tests harmful to it, with eventual result of its destruction.
Compared to that, the prospect of using it as biocomputer seems much better.

[stevebowers]

My daughter points out that a mind that didn't suffer would be useless, as it could never learn or react. We will just have to accept that our mind children will have bad times as well as good times.

[Dalex]

Your daughter is right. Back to the tissue.

The thing is that several critical problems would have to be solved first.
Namely:
1.Input/output interface
2.Nutrition

Brain in a vat is useless as computer if you can't provide it with data to process and way to express the results.
And if you biocomputer dies in few months it can hardly be cost-effective.

The input/output is in my opinion a much bigger problem. You would have to prevent scar tissue from forming around electrodes which is currently a major issue with all MMI in real life. Also the interface would need a way to translate analog input into digital output. Finally programing biocomputer is easier said than done.

[xilman]

The input/output is in my opinion a much bigger problem.

And in mine. Just about every bit of software I've created, whether running on hardware, other software or wetware, has needed more attention paying to the I/O than most everything else. I/O, whether from/to hardware, software or wetware, is hard to get right and usually requires several optimization cycles. Recently I've been trying to learn more about digital electronics through the use of FPGA dev-kits. Inputting from something as simple as a push button or outputting to as simple as a LED is distinctly non-trivial if all you've got is a blank page of Verilog and a few hardware data sheets.

It's no coincidence that the first thing to run on most every platform is "Hello world" or its equivalent.

[stevebowers]

The input/output is in my opinion a much bigger problem. You would have to prevent scar tissue from forming around electrodes which is currently a major issue with all MMI in real life. Also the interface would need a way to translate analog input into digital output. Finally programing biocomputer is easier said than done.

A recent article in New Scientist listed some possible methods to establish links between electronics and nerve cells; some of these require genetic or somatic modification to the interface cells.

[kch49er]

A while back i remember reading an article about rat brain matter being used to run a flight simulator and cyborged coackroaches, so we could control them via joysticks and the like..

edit Old but that was what I was thinking. Don't know what the latest research is


http://www.eurekalert.org/pub_releases/2...102104.php

http://neural.bme.ufl.edu/page12/page1/page1.html