Earlier this month, five researchers at the University of California, Berkeley, put out a paper discussing the possible development of mind-reading "neural dust," which could be implanted directly into the human brain to allow people to interact with machines.

The paper is what the MIT Technology Review calls a theoretical study: The idea is "littered with challenges beyond the state-of-the-art."

But according to the Berkeley team, this neural dust sprinkled into an individual's brain tissue could form an "implantable neural interface system that remains viable for a lifetime."

The particles of neural dust would be very small, not more than 100 micrometers across -- that's 100 millionths of a meter. Each particle would actually be a sensor capable of measuring electrical activity in neurons, covered in polymer to render it biologically neutral and backed by a piezoelectric material that could convert electrical signals into ultrasound. Thousands of these sensors could be constructed "at the tips of fine wire arrays," the paper explains. The wire arrays could then be inserted directly into brain tissue. Once the sensors pulled free of the wire, the arrays would be withdrawn.

Implanted in the brain, these sensors could theoretically work wonders. Ryan Whitwam of ExtremeTech speculates that they could enable brain-machine interfaces, mind reading or even "science-fictional telepathy."

The Berkeley team -- Dongjin Seo, Jose Carmena, Jan Rabaey, Elad Alon and Michel M. Maharbiz -- certainly have vision. The MIT Technology Review mentions that Maharbiz developed "the world’s first remotely controlled beetle," which the journal called "one of the top 10 emerging technologies of 2009."

But should the team try to make their proposed neural dust a reality, they face significant challenges -- including, rather critically, making sure the system is efficient enough to "avoid heating between skull and brain."

Earlier on HuffPost:

Loading Slideshow

• A Dinosaur Clone

  <strong>The Problem:</strong> We want dinosaur clones. Let’s be serious. Not even for "Jurassic Park" shenanigans, just to see what they really looked like. Bones, after all, only reveal so much about dinosaurs' actual appearance. One big limitation to dino cloning now -- <a href="http://www.nature.com/news/dna-has-a-521-year-half-life-1.11555">DNA degrades over time.</a> <strong>The Solution:</strong> To clone dinosaurs -- or indeed any extinct species -- we’d need a method of reconstructing DNA from the scraps we have. Go the further step of letting us custom-design DNA from the “scraps” of several animals, and this might help us increase the genetic diversity of low-population species -- or create entirely new breeds of animal. <strong>How Close Are We?</strong> On NPR earlier in 2012, futurist Stewart Brand talked about a new genetic reconstruction technique that could, over time, morph <a href="http://m.npr.org/news/Science/155717381">a contemporary elephant into a wooly mammoth</a>. Bran thinks this technique could work for other animals, and eventually allow for the rehabilitation of extinct species. South Korean and Russian scientists seem to be on the same track, and have recently moved forward in a bid to <a href="http://www.google.com/hostednews/afp/article/ALeqM5jFKsLZoidLt58zgUpmS3M-mAsFOQ?docId=CNG.c3a9e6de1510ab5f6807fbfeea405230.4c1">resurrect the mammoth by modifying the egg cells of an Indian elephant. </a>

• The Sum Of All Human Knowledge, In One Place

  <strong>The Problem:</strong> The Internet has been called the sum of all human knowledge. But it hasn't really earned that title. Most of the world's libraries haven't yet been scanned, let alone put on the web in computer-readable format, which means there are still huge swaths of knowledge that very few people can access. And as <a href="http://www.huffingtonpost.com/steve-berry/why-preserving-history-matters_b_1446631.html">libraries and museums lose funding and ancient manuscripts degrade </a>, we risk losing most of that knowledge forever, unless we find a way to save it. <strong>The Solution:</strong> Book scanning, like what <a href="http://books.google.com/">Google Books</a> is doing, is the first step. Next we'll need digitally readable text so the knowledge stored in this multitude of books can be extracted, sorted, compressed and disseminated efficiently. A technology for slowing or stopping the decay of old manuscripts would be a boon -- both for the sort of people who digitally archive books and for historians who will likely want to continue reading the books as primary source material. <strong>How Close Are We?</strong> Well, Google Books started scanning books from public libraries into its system in October 2004, <a href="http://www.huffingtonpost.ca/2012/10/04/google-and-us-publishers-_n_1939238.html">but has run into legal trouble with publishers for doing so.</a> Computer vision, especially in the field of image recognition and image restoration technology, <a href="http://nextbigfuture.com/2012/08/computer-vision-systems-can-now.html">continues to improve.</a>. And whether you know it or not, you've likely been helping in the digitization process as well. <a href="http://www.google.com/recaptcha/learnmore">CAPTCHAs</a> -- the boxes that one must fill in with a pair of words in order to access certains parts of websites -- helps computers transcribe difficult-to-read text from books.

• A Computer That Can Talk Back

  <strong>The Problem</strong>: You can try to talk to Siri, but <a href="http://www.huffingtonpost.com/2012/03/12/siri-lawsuit-apple-sued-misleading-commercials_n_1340025.html">it’s really nothing like taking to a real person</a>. Simple questions stump her, weird grammar boggles her, accents end her. A computer that could really talk to you -- well, that would be like a real personal assistant, albeit with the Internet as a brain. (Don’t think about that for too long.) <strong>The Solution: </strong>A computer that could actually understand human speech -- not just make startled attempts like Siri does -- would revolutionize web search, of course. It might even be able to solve more complex problems as well. Teach a computer to understand language and it could teach students basic writing skills, <a href="http://www.theatlantic.com/debates/education/">which are currently badly under-taught</a> -- <a href="http://www.huffingtonpost.com/2012/11/09/microsoft-voice-translator_n_2102158.html">or work as a no-cost universal translator. </a> <strong>How Close Are We?</strong> Siri continues to push the envelope on computer voice recognition, and <a href="http://www.slate.com/articles/technology/technology/2012/11/siri_vs_google_the_search_company_s_voice_recognition_program_gets_closer.html">Google Voice may be even better than Siri</a>.

• A Better Battery

  <strong>The Problem:</strong> <a href="http://www.wired.com/wired/archive/14.11/battery.html"> It’s a pity that even the best batteries on the market today are expensive, heavy and energy-sparse. </a> Consider this: The heaviest part of an electric car is the battery. The battery for the long-range Tesla Model S <a href="http://www.dailytech.com/Tesla+Model+S+to+Feature+17+Center+Console+Display+123+Instrument+Cluster/article20559.htm">weighs over 1,200 pounds</a> and <a href="http://www.teslamotors.com/no_NO/forum/forums/tesla-model-s-officially-rated-89-mpge-265mile-range">carries about 265 miles of charge</a>. That's 4.53 pounds of battery per mile driven, which is at least 3.53 pounds of battery too much. <strong>The Solution:</strong> We need a battery with better energy density. It’s as simple as that. For the sake of the world, or at least for the sake of our mobile phones. <strong>How Close Are We?</strong> Unsurprisingly, electric car manufacturers are very interested in this problem. <a href="http://www.technologyreview.com/news/506881/how-improved-batteries-will-make-electric-vehicles-competitive/">Toyota and GM have both funded research into solid-state batteries, while independant startups have grown more inventive</a>, creating combinations that could potentially replace lithium ion -- magnesium ion, for instance, or nuclear batteries. Still, no silver bullet has emerged.

• An Automated Checkup

  <strong>The Problem:</strong> Medicine is hard. Taking care of yourself is hard. We’re all busy, we’re all stressed -- and around the holiday season, we’re all sick. Checkups come once, maybe twice a year, and in the meantime all sorts of bad things can happen to us. Wouldn’t it be nice to nip some sicknesses in the bud beforehand? <strong>The Solution:</strong> Here’s an idea: a camera (or something like a camera) that could scan your body, maybe check your blood and urine and report back to you what health problems you might be at risk for -- and what you can do to stop yourself getting sick. Not enough Vitamin D? Drink more milk. Sinuses getting spongy? A cold’s coming on. Mole changing color? Probably want to get that checked pretty soon. It’s not a substitute for checkups, of course. But it’s better than self-diagnosing via <a href="http://www.webmd.com/">WebMD.</a> <strong>How Close Are We?</strong> Well, automated blood analyzers and <a href="http://www.amazon.com/s/ref=pd_sl_1h8xdou6zj_b?rh=i%3Aaps%2Ck%3Aurine+testing+strips&keywords=urine+testing+strips&ie=UTF8&psrk=urine+test+strips">urine test strips</a> are already well a part of medical science, though <a href="http://www.alibaba.com/showroom/automated-blood-analyzer.html">the price of automated blood analyzers might be prohibitive for average buyers.</a> And between full-body MRI, CT scans, PET scans and ultrasounds, we probably have the technology to make something like the “camera” part of the machine -- but it would be prohibitively expensive, relatively radioactive, and huge. Decrease the scale, price and radioactivity of a lot of already-extant inventions, and you’ve got this machine.

• A Cure For Distance

  <strong>The Problem:</strong> In 1964, Japan built the first bullet train. Since then, transportation has only gotten worse. Yup. <a href="http://engineering.mit.edu/live/news/188-why-hasnt-commercial-air-travel-gotten-any-faster">Commercial airflight is now slower than it was in the 1960's</a> -- and that's not counting the ever-longer security lines. Car travel hasn't gotten much faster since the dawn of the interstate highway, our railroads have barely changed in 50 years, and while supersonic jet engines exist, they have yet to become used for mass transport. No readily available flying cars, no practical jet pack solutions, no teleportation -- not even commercial flights that could take us from New York to San Francisco in less than four hours. <strong>The Solution:</strong> Anything that can dramatically reduce transportation time to below that of commercial air flight, and be scaled up for massive and inexpensive use. We're not even going to propose a single solution here -- we just need something that can tackle physical distance better than we’ve been doing. We live in a globalized world but we're still stuck using transportation from almost 50 years ago. <strong>How Close Are We?</strong> A lot of solutions have been proposed, from <a href="http://www.huffingtonpost.com/2012/09/25/hyperloop_n_1913683.html">Elon Musk’s hyperloop</a> to the <a href="http://www.bbc.com/future/story/20120813-flying-cars-ready-for-take-off">much-attempted flying cars.</a> But nothing, so far has been implemented, or even gotten much private or public attention.

• Collaborative Lawmaking

  <strong>The Problem:</strong> A lot of people have said <a href="http://www.huffingtonpost.com/2011/03/29/internet-polarizing-politics_n_842263.html">the Internet increases polarization</a>. Liberals and conservatives, the argument goes, each hide in their own little corners of the web, not really interacting with each other in any meaningful way. <strong>The Solution:</strong> There are places on the Internet where bipartisanship works far more smoothly than it does in Washington. Wikipedia, for instance, now has <a href="http://en.wikipedia.org/wiki/Fiscal_cliff_debate">a page on the 2012 Fiscal Cliff Debate</a> that has a note on top. “This article needs consensus. Please discuss any content disputes and controversial issues on the article's talk page,” the note reads. <a href="http://en.wikipedia.org/wiki/Talk:United_States_fiscal_cliff">Move to the talk page,</a> and there’s a miracle of bipartisan discussion on exactly how the Fiscal Cliff should be discussed, with civility that Washington (or the rest of the internet) should perhaps envy. And if Wikipedia can do it, why can’t other parts of the Internet? If an encyclopedia can be open-sourced, why can’t other important documents? It wouldn’t take much to “invent” collaborative lawmaking -- just a program that lets people look at drafts of laws, suggest amendments and tweaks, and perhaps find the consensus that Washington can’t. Surely there are disability laws and tax laws that Republicans and Democrats can agree on -- especially if the people reaching consensus don’t need to worry about re-election campaigns, fundraising or toeing the party line. <strong>How Close Are We?</strong> There are no technical obstacles to collaborative lawmaking, but no one’s started an initiative to put it into action either. The closest we’ve come is a <a href="http://www.wired.com/threatlevel/2012/09/patent-busting-crowdsourced/">collaborative effort by Stack Exchange and the US Patent Office to crowdsource patent applications and prevent patent trolling.</a>