Tag: MIT

MIT Researchers Use Reddit To Create An AI That Only Thinks About Murder

Artificial Intelligence is one of those terms that induces a polarising reaction from different people. Some believe it’ll make our lives easier and others believe we will soon be an episode of Black Mirror thanks to AI. For tech companies, however, Artificial Intelligence is their new favourite phrase. Google, for example, probably set the world record for using the phrase “AI” the most number of times in an hour at Google I/O 2018. Ever since the Pixel 2 devices were launched, Google has not shied away from using this term in almost everything it does.

While AI assistants like Alexa, Google Assistant do make our lives easier to a certain extent, nobody wants a psychopath AI in their lives. Anyone who has watched Altered Carbon can imagine how creepy and intrusive AI can be if we keep making advancements in the field without keeping a check on it. Researchers at Massachusetts Institute of Technology (MIT) unveiled their new creation, a disturbed AI named Norman. Yes, this AI is named after the “lovable” character from the 1960 film, Psycho. The researchers write:

Norman is an AI that is trained to perform image captioning, a popular deep learning method of generating a textual description of an image. We trained Norman on image captions from an infamous subreddit (the name is redacted due to its graphic content) that is dedicated to document and observe the disturbing reality of death. Then, we compared Norman’s responses with a standard image captioning neural network (trained on MSCOCO dataset) on Rorschach inkblots; a test that is used to detect underlying thought disorders.

Essentially, this AI has very disturbed responses when compared to a general AI. While the Rorschach test has its own doubters whether it is a valid way to measure a person’s psychological state, Norman’s responses don’t need a test to be labelled creepy. The image below captures the level of effect that Reddit thread had on Norman’s ability to perceive images.

The researchers have said that the aim of the experiment was to show how easy it is to bias any AI if you train it on biased data. This experiment raises some telling points about AI and its rapid advancements. Google recently came under scrutiny after it demoed the Google Assistant fooling human beings in a telephonic conversation. Google has since conceded that when this feature rolls out, the Google Assistant will inform the person on call that they’re talking to Google Assistant.

June 8, 2018
“Smart Stickers” Which Will Detect And Prevent Sexual Assault

Manisha Mohan, an Indian grad student and researcher at the MIT, invented “Intrepid” – an electronic sticker that can detect forcible grabbing. The smart clothing detects “forcible touch” and sends a distress signal to your network of friends and family via message or a phone call.

The Intrepid can be attached to any piece of clothing and connected to user’s phone via Bluetooth. If the piece of clothing attached to the device is taken off forcibly, the device waits for 30 seconds after which a “loud alert” goes off for 20 seconds. If there is still no response from the wearer, an SOS is sent to the listed emergency contacts. With the help of an accompanying app, Intrepid can also record encounters in the case of no timely response from the wearer. These recordings can be further used in the court in the case of legal proceedings.

[embedyt] https://www.youtube.com/watch?v=RgU_evcSPrs[/embedyt]

The “Smart Stickers” operate in two modes – active and passive. Manisha explains that, “An active mode [for] for instances when the victim is unconscious or cannot fight against the assaulter, for example in the case of infants, bed-ridden patients, elderly, disabled, intoxicated people and the passive mode where the victim can self-actuate the safety mechanism.”

The Intrepid uses methods like olfactory stimuli (sense of smell) to detect initial signs of assault. It differentiates between the effect of smells on sexual arousal and partner selection with the effect of smells on forced touch or disrobing.

Sexual abuse, assault, and harassment are regarded as some of the most common human rights violations in the world. As per the data release by the NCRB and Delhi Police, cases of sexual assault across the country remain high, with India’s “rape capital” Delhi registering six cases of rape on an average every day. The reports paint a pretty bleak picture in regard to ensuring safety against sexual assault.

With crimes against women on the rise, and women being left with no choice but to stay indoors after dark, a device like Intrepid can not only help by improving the response time in cases of sexual assault but more importantly, it can make women feel safer when stepping out on the streets.

The proposed solution, as Manisha states, can help combat rape, assault, harassment, child sexual abuse, abuse of elderly and disabled and might offer some peace of mind.

August 4, 2017
Self-Charging Batteries to End Smartphone Battery-Life Woes

We have seen some incredible breakthroughs in technology lately but one issue that still needs to be solved is that of battery trouble. No matter how fancy a gadget gets, it is almost always limited by its battery life. The next big thing the world of tech requires is never-ending batteries, and looks like MIT is paying attention.

A group of researchers at the university are working on self-charging batteries, which in turn means never-ending battery life. The idea is to produce electrical energy out of the mechanical energy of locomotion. The mechanical energy produced as a result of walking can be used to charge electronic devices.

The procedure involves two thin sheets of lithium alloys to serve as electrodes with a porous sheet of polymer, after it has been soaked in liquid electrolyte, placed in between. This design is efficient at moving lithium ions, while at the same time being attached to sensors on the joints on the body like knees or wrists. The sensors would absorb mechanical energy of locomotion and convert it to electrical energy.

This tech could also be used to charge wearables like smartwatches, along with phones. However there is still a long way to go. Right now, the use of this technique can achieve 15% efficiency for a gadget at best. The research team hopes to increase this efficiency to a 100% soon.

January 7, 2016
MIT Successfully Manages to Make 3D Printed Glass

A new development at MIT has shown us how to expand the 3D printing technology and use it to make 3D printed glass.

A method called G3DP is said to allow the creation of complex 3D glass structures to be printed in a similar fashion to plastic constructs. This process uses two chambers – one that acts as a kiln cartridge and the other that works to melt structures together. The molten glass is then distributed through an alumina-zircon-silica nozzle that pours the material out smoothly.

The entire process was developed by a team including the MIT Media Lab’s Mediated Matter group, the MIT Glass Lab, the Wyss Institute and MIT’s Mechanical Engineering Department. This new breakthrough is also said to advance the process of creating fiber optic cables that transmit data more efficiently. One of the project’s leaders, Neri Oxman, while conducting research said :

“Now [we can] consider printable optoelectronics, or the possibility of combining optical fibers for high-speed data transmission by light, combined within glass printed building facades.”

One of the research’s vision is to one day employ this technology in building facades. However, reports suggest that the entire process is slow and therefore not efficient enough, but it can still create complex structures for other purposes.

There has been no information regarding the implementation of this device by any commercial application, but the MIT G3DP is going to be showcased at an exhibit in the Cooper Hewitt, Smithsonian Design Museum in New York next year.

August 24, 2015
Elon Musk Cautions Against Creating Artificial Intelligence

In the knowledge and information age that we live in now, a great amount of research is targeted towards creating Artificial Intelligence (AI). The goal is to create a computing system which is as sentient as a human being with same levels of intelligence. But a barrage of people have warned against the creation of such a machine. Adding to the list is Tesla and SpaceX CEO Elon Musk.

At MIT’s annual AeroAstra Centennial Symposium, Elon Musk said that developing an artificial intelligence maybe akin to summoning the demon. He explained, “You know those stories where there’s the guy with the pentagram, and the holy water, and he’s like – Yeah, he’s sure he can control the demon? Doesn’t work out”.

He said that there is an immediate need for a regulatory oversight over the efforts of AI development. “I’m increasingly inclined to think there should be some regulatory oversight may be at the national and international level, just to make sure that we don’t do something very foolish.”

Musk also warned that HAL 900 is a Puppy dog in comparison to what humans can create

He went ahead to call it the biggest existential threat to humanity and added that the HAL 9000 from 2001 A Space Odyssey would look like a puppy dog in comparison to what humans can build.

As a trusted name in the field of scientific research, this warning is concerning for folks who were dreaming of incorporating a JARVIS-like system in their house someday. Movies like Terminator, Matrix and many others have warned us about the consequences of AI taking over human civilization. A system that can get instant access to all of humanity’s knowledge in the first couple of seconds of its existence may indeed revolt against its imperfect human masters and overpower them. This is one cautionary tale that should be taken notice by not just the stakeholders but the entire population. We might inadvertently lose the alpha predator tag to our own creation and become its prey.

October 28, 2014
MIT’s THAW Brings Ultimate Seamless Interconnectivity Between Smartphones and PC

With a PC and Mobiles being an absolute necessity, it is necessary to have seamless interaction between them. Apple’s latest ecosystem attempts to do just that. Continuity features like Handoff let users continue their work from their phone to their Mac PC’s without a delay. Not to be left behind, Google also announced some interconnectivity features between Android desktops and Chrome systems. But these attempts are still futile in comparison to MIT’s THAW which takes seamless interconnectivity to a whole new level.

Two teams of researchers at MIT have developed a smartphone system called THAW which allows users to transfer files through and fro from PC to a smartphone. This can be done simply by touching the phone to the screen of the desktop PC. You can take the webpage you were working on your phone by just touching the phone to the screen. It almost feels like those screens in sci-fi flicks which can be shifted from one display to another, like Avatar.

THAW makes you phone a physical handle through which you can send or receive files directly from the screen. The secret behind the technology is the phone camera. The camera takes the image from the desktop screen and then through software manipulations the files are accessible on the phone. This feature will help solve the file management clarity issue that everyone is trying to address. Apple’s Cloud Storage now gives the ability to change the content, and it will appear the same from all your devices. THAW just takes it a bit further.

Now this project is still in its infancy and may or may not come to the market. But just the knowledge of the possibility of this technology will create a boom for devices with such functionalities. This looks like an excellent feature to have and looks so simplistic and effortless. Hopefully, the companies are watching; this is one feature that everyone will love. MIT researchers claim that this can work with practically any smartphone with accelerometer, so come on Samsung, here’s an idea for the Galaxy S6.

September 20, 2014