AI芯片植入人脑:改善情绪并使注意力更集中

2017 年 11 月 29 日 全球人工智能


——免费加入AI技术专家社群>>

——免费加入AI高管投资者群>>

——日薪5k-10k招兼职AI讲师>>

--全国招募1000名AI推广大使>>

据外媒报道,美国国防部高级研究计划局(DARPA)正在研发一款可以植入人脑的人工智能芯片,可以通过算法监测大脑中的情绪障碍,并利用电子脉冲改变大脑的化学反应,从而实现深层次的脑部刺激,使大脑自动恢复健康状态。另外,有专家认为芯片植入有助于监测帕金森、抑郁症、癫痫等大脑、神经疾病的发作,实现健康跟踪;以及帮助人类集中注意力。但是对大脑信息的探索也面临着伦理道德方面的风险。

The US military has begun testing AI brain implants that can change a person's mood on humans.

These 'mind control' chips emit electronic pulses that alter brain chemistry in a process called 'deep brain stimulation.'

If they prove successful, the devices could be used to treat a number of mental health conditions and to ensure a better response to therapy.

The chips are the work of scientists at the Defense Advanced Research Projects Agency (DARPA), a branch of the US Department of Defense which develops new technologies for the military.

Researchers from the University of California (UC) and Massachusetts General Hospital (MGH) designed them to use artificial intelligence algorithms that detect patterns of activity associated with mood disorders.

Once detected, they can shock a patient's brain back into a healthy state automatically.

Experts believe the chips could be beneficial to patients with a range of illnesses, from Parkinson's disease to chronic depression.

Speaking to Nature, Edward Chang, a neuroscientist at the University of California, said: 'We’ve learned a lot about the limitations of our current technology.

'The exciting thing about these technologies is that for the first time we’re going to have a window on the brain where we know what’s happening in the brain when someone relapses.'

The chips were tested in six people who have epilepsy and already have electrodes implanted in their brains to track their seizures.

Through these electrodes, the researchers were able to track what was happening in their brains throughout the day.

Older implants are constantly doing this, but the new approach lets the team deliver a shock as and when is needed.

By tracking a patient's brain activity over the course of one to three weeks, they were able to create an algorithm to ‘decode’ their moods.

The MGH team then discovered that by delivering shocks to regions of the brain that deal with decision-making and emotions, participants performed significantly better in set tasks.

This included matching images of numbers or identifying emotions on faces.

Difficulties with concentration and problems with empathy are characteristics of a range of mood disorders.

Although researchers won't be able to read people's minds, the chips do raise a number of ethical concerns.

Alik Widge, engineering director of the MGH team, added: 'We will have access to activity that encodes their feelings.'

Dr Widge’s group is working with neuroethicists to address the moral implications of their work.

The full findings of the study were published in the journal Nature.

http://www.dailymail.co.uk/sciencetech/article-5120441/Mind-control-AI-chips-alter-persons-moods.html


热门文章推荐

招聘|AI学院长期招聘AI课程讲师(兼职):日薪5k-10k

浙大才女:用人耳听不到的超声波(攻击)控制语音助手!

Science:最新发现哈希可能是大脑的通用计算原理!

厉害|波士顿动力新版人形机器人Atlas,后空翻很完美!

吴恩达:AI论文已经够多了,赶紧“搞点事”吧!

厉害了!吉利宣布收购飞行汽车公司Terrafugia !

周志华:实验表明gcForest是最好的非深度神经网络方法

黑科技|Adobe出图象技术神器!视频也可以PS了!!

史上第一个被授予公民身份的机器人索菲亚和人对答如流!

浙大90后女黑客在GeekPwn2017上秒破人脸识别系统!

周志华点评AlphaGo Zero:这6大特点非常值得注意!

登录查看更多
0

相关内容

强化学习最新教程,17页pdf
专知会员服务
174+阅读 · 2019年10月11日
[综述]深度学习下的场景文本检测与识别
专知会员服务
77+阅读 · 2019年10月10日
《科学》(20190517出版)一周论文导读
科学网
5+阅读 · 2019年5月19日
《科学》(20190426出版)一周论文导读
科学网
5+阅读 · 2019年4月27日
人工智能 | SCI期刊专刊信息3条
Call4Papers
5+阅读 · 2019年1月10日
推荐|深度强化学习聊天机器人(附论文)!
全球人工智能
4+阅读 · 2018年1月30日
【推荐】深度学习情感分析综述
机器学习研究会
58+阅读 · 2018年1月26日
计算机类 | 期刊专刊截稿信息9条
Call4Papers
4+阅读 · 2018年1月26日
计算机视觉近一年进展综述
机器学习研究会
9+阅读 · 2017年11月25日
Arxiv
92+阅读 · 2020年2月28日
Arxiv
19+阅读 · 2019年4月5日
Arxiv
7+阅读 · 2018年1月31日
VIP会员
相关VIP内容
强化学习最新教程,17页pdf
专知会员服务
174+阅读 · 2019年10月11日
[综述]深度学习下的场景文本检测与识别
专知会员服务
77+阅读 · 2019年10月10日
Top
微信扫码咨询专知VIP会员