3 Emerging Future Technologies That Will Make You Smarter

3 Emerging Future Technologies That Will Make You Smarter

We’ve all wished at one time or another that we were smarter. Maybe it’s so that you can ace that next math or physics test, or perhaps you’re just sick of losing at Trivia Crack. Whatever the reason you’d like to be smarter, read on to learn more about some future technologies that have the potential for making us all smarter human beings.

1. tCDS: transcranial Direct Current Stimulation


This technology is still in it’s infancy, but there are currently some companies selling units online. Reddit user Ryantific_theory does a great job explaining more about it:

tDCS (transcranial Direct Current Stimulation) has shown significant improvements in learning speed in a few scientific studies. To sum it up, it functions by running a weak DC (non-alternating) current across cortical regions, and producing a persistent excitability in the affected neurons, this is known as LTP (Long Term Potentiation) and is one of the hallmarks of learning. Anyway, it has potential to produce literal thinking caps, and some people have already started selling basic units for it.”


So basically if you’re willing to slap some electrodes on yourself every day (or every couple of days) and zap your brain with electrical current, you could increase your intelligence without doing anything else. Want to take a look at some units currently for sale? They can be found on the following websites (LineMind.com does not have any affiliation with these providers):





There’s even a do-it-yourself option for those that want to get smarter on a budget.

Ryantific_theory goes on to explain that

“they’re all fairly early stage, and somewhat crude overall. While there’s nothing to stop you from living your life however you want, I would be somewhat cautious. If it is something that you decide to do, I would look up the lengths of time studies had it running on subjects to get an idea of what’s reasonable or crazy. The review article refers to stimulation times of 9-13 minutes producing long lasting plasticity changes, and those can be maintained by stimulation once a day.

The danger in overstimulating, is that you can force neural consolidation if you apply the current long enough, and it would lock in whatever neurons happen to be in the current path with unpredictable, if relatively minor, consequences. So just say no to overstimulation.”

2. Neuromodulation: Deep Brain Stimulation


Deep brain stimulation involves implanting a “brain pacemaker” directly into the brain. This brain pacemaker would then send electrical impulses to different regions of the brain, thereby influencing how the brain responds. It’s current applications are all for medical reasons like Parkinsons disease, major depressive disorder and obsessive compulsive disorder among a few others. According to Ryantific_theory:

“Neuromodulation (Deep Brain Stimulation) will become massively important as the implementation reaches continually higher and higher resolutions, and is applied beyond Parkingson’s and the few other select conditions it’s currently used for. Conceptualize the fact that everything you are exists as a series of electrical impulses in your skull, as we decode more and more of the neural code, we will eventually create implants that give us an unprecedented control over our own mind.

Ethically, this is a major issue, because if you can turn motivation, happiness, satisfaction, desire, and interest on and off at will, where do we draw the line? For people struggling with anxiety or depression, or anything they don’t like, we will have the technology to allow people to fundamentally change themselves. Stimulation of the hippocampus has been shown to also potentially help improve memory formation and recall.

Ultimately the idea of an implanted neural net with a number of connections to critical areas seems likely, but directly offloading memories and other sci-fi tropes is deep, deep future. DARPA happens to be working on a internal display that takes advantage of the accessability of the V1 cortex, and exact visuo-spacial mapping to put a “perceived” display into visual space, which basically amounts to a screen shown on your internal representation of visual space. Normally I’d write it off as another far flung hope, but they’re actually developing tech to make it possible, and recently developed with the U of Wisc a graphene optoelectric stimulating recorder. Which is sweet.”

Although he also states applications that would make a person smarter are perhaps 20 years away and I know you need help with Trivia Crack sooner than that. I also think it’s worth reiterating a striking comment from above:

Conceptualize the fact that everything you are exists as a series of electrical impulses in your skull.”

You know that hoarder cat lady down the street? Or what about that really ugly sweater your grandma thought you would love for Christmas. It’s all a result of a series of electrical impulses in the brain. Damn, things sure do easily go awry.

3. Changes in Education and Access to Education

education technology

The final future technology that will make us smarter won’t necessarily be technology itself, but rather it’s application in the education process. As Redditor Chispy stated:

“Virtual and augmented reality will allow for cheap and efficient methods of visual learning. Couple it with repetitive learning strategies, and you can easily ‘Gamify’ Education, and make learning happen through aspiration and collaboration.”


It doesn’t matter how many times someone tells me how to do something, I learn by actually doing it myself. Being able to insert myself into a virtual reality scenario that allowed me to complete tasks would dramatically increase how quickly I learned. Plus, it would be a lot more fun and engaging.

As another Redditor, sprocket86 put it:

“The technologies that will improve collective human intelligence are the ones that spread the most knowledge. Education is the bottleneck in society’s intelligence, not brain power. If you’re looking for the most promising emerging technologies, I say it’s language/teaching software. AI teachers can engage in conversation and efficiently find answers and give you a personalized education. This would basically require Turing-test-level AI if you’re taking class for fiction writing, philosophy, or something highly abstract, but many classes are based on fact retrieval and explanations. Imagine an interactive Khan Acadamy for anything and everything you want to learn about. Then you can start playing around with brain enhancements”


Can you imagine how much smarter the collective intelligence of everyone on the planet would be if we all had access to a teacher created and run on artificial intelligence? It would be like allowing everyone to have their own tutor.

In conclusion, most of the future technologies that will make you smarter are a ways off. According to Ryantific_theory:

“In the short term (next 15ish years), I’m not sure that there will be a huge impact in making people smarter. Almost everything being done now in the field is medically motivated or highly experimental, and as a result the first people to benefit will be those suffering from neurological conditions, ataxia, seizures, Parkingson’s, Alzheimer’s, dystonia, and the like. It’s not that we don’t want to make people smarter, it’s just that the requisite understanding to really help with that lays well beyond the solutions to current neural problems.

Likely the biggest impacts will have more to do with scientific optimization, developing and determining the best methods for people to learn, coupled with the massive spread in access to (factual) information should have a significant impact on the average level of education. That alone will be huge, if more social than technological.”

He goes on to say that:

“in the future that’s within sight, at this moment, the best path I can see to making humans smarter stems a lot more social factors than technological. But the cool tech factors that will help will likely involve direct stimulation of neural tissue. The problem mainly lies in the fact that from a neural standpoint we don’t have a complete picture of what makes smart people smarter than less smart people, and we can’t effectively experiment until we know what we’re trying to do.

So the biggest impact to making humans smarter will probably have more to do with changing how effective we are at learning, rather than just plugging in BRAINS++ and magically making people smarter.”

Even though most of the benefits of these technologies won’t be seen in the near future, they’re still fun to think about. Although I’m also not sure I’d be too keen on having an implant put in my head, but I suppose if everyone else was doing it….