BrainGPT: Mind-Reading Technology Turns Thoughts Into Text

BrainGPT

Mind-reading has long been the stuff of science fiction stories. When it comes to technology, however, the line between fiction and reality is becoming blurred.

Scientists at the University of Technology Sydney have begun exploring that line. 

Recently, they’ve taken some big leaps forward with mind-reading technology… And it’s mind-blowing!

They hope to create a future where they will be able to give a voice to those who have lost theirs to medical and neurological issues. 

The new Artificial Intelligence, called DeWave or BrainGPT, has successfully predicted thoughts with a 40% accuracy.

The goal is to reach 90%.

Does Mind-Reading Technology Exist?

Despite all the recent innovations in artificial intelligence, there is no technology, so far, that can open up the mind and read its contents like a diary. There is, however, technology that can study the human brain and glean all sorts of information. This is what is meant by the term “mind-reading technology”.

Advancements in technology can tell researchers how illness or injury impacts the brain. Scientists can learn how the brain responds to certain sights or sounds. Using this information, they can use Artificial Intelligence to help them decode patterns and “read minds.”

And now with BrainGPT, these patterns can be turned into text.

Mind-Reading Technology Can Create a New Era of Communication

For the first time in history, spinal injuries are becoming something that doctors can reverse. Although it is a long healing process, scientists have made amazing progress in treating paralysis using Artificial Intelligence (AI).

Those with paralysis or stroke often communicate through eye movements or twitches. Imagine a world where mind-reading technology can allow us to begin understanding and even communicating with people in this condition.

Mind-reading technology could also be used to understand Organoid Intelligence (OI). This is an emerging field of study where scientists are creating a bio-computer using human brain cells. By gaining a better understanding of how the human brain functions and responds, scientists can begin to answer important questions as OI progresses.

• Can brain organoids feel sensations?
• Do they show signs of communication?
• How do they respond to certain stimuli?
• At what point, if any, do they begin functioning like a real brain?

What Is Mind-Reading Technology?

Any technology that investigates what’s happening in the human brain could be called mind-reading technology. 

Take Algorithms, for example. Algorithms are used in browsers to enhance the user’s experience. They determine what a user needs or would like to see based on previous activities and interests. Another would be an email program that can finish a phrase or sentence before the user does. 

These kinds of algorithms are based on observations of past behavior patterns. They are data. This data can be used to calculate the likelihood of what action or choice will follow. 

DeWave is an AI that is utilized by BrainGPT. BrainGPT is a language model that also uses vast amounts of neuroscience data to help researchers. This technology is reaching a point where it’s using data to read minds in a very real way. 

How Does Mind-Reading Technology Work?

For mind-reading technology to work, it’s first important to understand how the human brain processes language.

Words don’t exist in the human brain the way they exist in writing or speech. The brain manifests words in the form of brain waves, or electrical impulses, that ignite when a word is spoken or read. These impulses are unique and occur in a wide variety of places in the brain. 

In tests for the DeWave software, subjects wore a snug-fitting cap. The device took a reading of their brain waves as they read silently from assigned material. The DeWave technology then studied these brainwaves. Through this process, DeWave learned to associate each wave with a specific word. The result is a kind of dictionary of brainwaves that allows DeWave to interpret what a subject is thinking.

At 40% accuracy, it’s not perfect. However, the goal of 90% sounds pretty incredible. Scientists are learning more about why inaccuracies occur with DeWave, and how it can be advanced to make BrainGPT more accurate.

Why Isn’t BrainGPT Perfect?

Technologies like BrainGPT and DeWave have many Inaccuracies that are caused by several things. 

• Reading vs. Thinking: Research teams collect data while subjects read printed words. As people read, spaces between words and certain punctuation marks signal them to take a pause. But when people think freely, such pauses don’t exist. This can make it challenging for BrainGPT to recognize words and sequences.

• Clustering Words: The brain tends to cluster words with a similar meaning. This causes them to have very similar brainwave patterns. This forces the BrainGPT to make a choice. Sometimes it will replace a common word with something similar, but different. For example, “The author” might be interpreted as “The man”.

• Brain Differences: Scientists are still studying how the brain processes language in many different situations. For example, people who speak English as a second language may have a very different set of neural responses than native speakers. Researchers of mind-reading technology have yet to understand the impact of various accents and other speech differences.

For now, scientists are excited that the technology has become accurate enough to interpret much of what a person is thinking. Word for word, literal thought interpretations are still a thing of the future.

However, as scientists work toward this goal, dire questions have arisen about the safety and ethics of this new technology.

How Should We Handle Mind-Reading Technology as It Evolves?

The thought of allowing technology to peer into our most private minds is frightening for a lot of people. It has both amazing potential and risk.

The technology could give a voice to people who can’t speak. There has also been speculation that such technology could be used to identify people with serious mental health problems. This could give the opportunity to provide help for those who need it at a time when it matters.

Mind-reading technology can also be used to develop a lie detector that is impossible to cheat, and even prevent crimes before they happen. 

However, this technology could have some serious downsides. It could be used to steal highly secure information. This could include passwords to computers containing financial or medical records. Results can also be misinterpreted which can lead to false judgments.

Two approaches have been suggested by researchers for dealing with the ethics of mind-reading Technology.

• Embedded ethics: This approach would involve programming ethics into the software and hardware involved in these mind-reading devices. This ensures the technology is incapable of crossing boundaries laid out for it.
• Adversarial ethics: This would involve the development of laws and regulations. These limits would be enforced by authorities governing the way this technology is used. 

Ethics and rights will need to be evaluated as neuroscience and Artificial Intelligence evolve. Organizations like The Neurorights Foundation are already planning for the future. They’ve identified five different areas of specific focus:

• mental privacy, personal identity, free will, fair access to mental augmentation, and protection from bias.