A.I. Just Designed An Enzyme That Eats Plastic
Вставка
- Опубліковано 6 чер 2024
- » Podcast I Co-host: / @throughtheweb
» ColdFusion Discord: / discord
» Twitter | @ColdFusion_TV
» Instagram | coldfusiontv
-- About ColdFusion --
ColdFusion is an Australian based online media company independently run by Dagogo Altraide since 2009. Topics cover anything in science, technology, history and business in a calm and relaxed environment.
» Facebook | / coldfusioncollective
» Podcast Version of Videos: open.spotify.com/show/3dj6YGj...
podcasts.apple.com/us/podcast...
ColdFusion Music Channel: / @coldfusionmusic
ColdFusion Merch:
INTERNATIONAL: store.coldfusioncollective.com/
AUSTRALIA: shop.coldfusioncollective.com/
If you enjoy my content, please consider subscribing!
I'm also on Patreon: / coldfusion_tv
Bitcoin address: 13SjyCXPB9o3iN4LitYQ2wYKeqYTShPub8
-- "New Thinking" written by Dagogo Altraide --
This book was rated the 9th best technology history book by book authority.
In the book you’ll learn the stories of those who invented the things we use everyday and how it all fits together to form our modern world.
Get the book on Amazon: bit.ly/NewThinkingbook
Get the book on Google Play: bit.ly/NewThinkingGooglePlay
newthinkingbook.squarespace.c...
Sources:
www.theguardian.com/environme...
utw10252.utweb.utexas.edu/peo...
www.nature.com/articles/s4158...
www.forbes.com/sites/davidrve...
www.chemistryworld.com/news/p...
utw10252.utweb.utexas.edu
My Music Channel: / @coldfusionmusic
//Soundtrack//
Bear Face - MXTHERBXY
No Spirit - Reminiscing
Owen - Bad News
Edward Sharpe and the Magnetic Zeros - Life Is Hard (Teen Daze Remix)
Sean Williams - Fray
Burn Water - Distant Dreaming
» Music I produce | burnwater.bandcamp.com or
» / burnwater
» / coldfusion_tv
» Collection of music used in videos: • ColdFusion's 2 Hour Me...
Producer: Dagogo Altraide - Наука та технологія
I've recently started getting into computational chemistry, and I have to say, it is some of the wildest stuff I've studied recently. Seeing how machine learning is applied into pure chemistry is one thing. But I really want to see how ai and machine learning can be applied to chemical engineering, especially reaction engineering. Finding the optimal reaction conditions to maximize output and yield, minimize costs and operation time, it's not an easy task.
I'm really looking forward to where this goes, especially in catalysis.
The UA-cam channel "Two Minute Papers" shows how fast AI training is going for visual computing.
This whole AI eco-system is evolving lightning fast, and for sure will speed up other science platforms!
@@N0N0111 pretty soon AIs will be designing fully functioning, hyperfit, hyperintelligent and fully controllable creatures that will replace humans in many laborious, dangerous, and most vital tasks that they perform in their day to day... uh oh..
@@N0N0111 I'm a big fan of 2 minute papers! They are great at breaking down complex phenomena for non-experts! And I'm seeing it firsthand in physics and chemistry, because that's my line of work. It'll be cool to see AI's work in other fields as well, definitely!
A lot of people seem to worry that AI will replace humans, but I think that a combination of human and AI work will be able to achieve more than that which one of the two could do alone.
Have you come across Lee Cronin and the Chemputer? He has an interview with Lex Fridman!
@@BrownCreature I have not. I'll have to check it out!
So many of the comments missed a few key points- (1) the AI optimized the enzyme, not the bacteria, so no replication(gray/grey-goo) issue. (2) The enzyme still requires temperatures outside of ambient temperatures to break down the PET within a week, so it’s not going to destroy our infrastructure if it gets out. (3) Yes, CO2 is a byproduct, but far less is emitted by this breakdown than burning the plastic.
Freon was the safest gas you could use for A/C systems when it was invented. Doctors used to tell pregnant women to smoke because it was good for the baby. Mercury, Asbestos, Roundup.... Any of these ring a bell? It's not a bad thing to have a bit of healthy skepticism, especially if you have been around long enough to remember the other many lies about great inventions of the past. Least we not forget about what happened the last time something escaped from a lab since we are literally still dealing with it.
Thanks for this summary! Clears up alot
How do we manufacture the enzyme if not via engineered bacteria...?
Just going to point out that point 3 is chemically impossible. The amount of CO2 emitted is exactly the same, it’s a matter of how much time does it take to reach the atmosphere.
@@theondono Am I wrong in my understanding that the enzymatic breakdown of the polymer results in individual monomers and CO2? There’s still plenty of carbon left in the monomers that could be converted into CO2 if the plastic was completely burned…
50 C doesn't seem that hot. Lots (though not most) of places could easily get that temperature without added energy for several months a year just by building a structure meant to magnify the ambient heat. Heck, my car gets to 50 C if I leave the windows up.
It's still a temperature much much lower than ambient, except for the hottest months in some of the hottest regions on Earth.
@@iCarus_A I think you mean higher
@@OatmealTheCrazy yes
I thought the same. 50 degC is a very reasonable operating temperature. In composting, the pile is sometimes heated to 50 degC to make sure the thermophilic bacteria degrade the food waste faster. Plastics are currently recycled by melting then cracking them into monomers at temperatures much higher than 100 degC.
All you need to do is have the processing center in a greenhouse in Arizona. Guaranteed 50°C/112°F temperatures for at least 9 months out of the year. And it wouldn't need too much help even in winter.
so nice of you to have the sources in the description! nice video ty
Science has come so far. It's hard to choose which is more amazing, the fact that we can decompose plastic or an A.I designed the enzyme.
How far has "science" come? How is your life improved? How has it made you more complete? And what do you even think science is?
I love it. The only thing better is holding those companies responsibke for what they did
Both
@@philindeblanc how far has science come? Look around you man
@@philindeblanc for starter, youre commenting today on this video because of science..
What products of science are available to you and how you use them depends on you.
It would be interesting to hear about the potential health issues micro plastic is causing
well,I got a goat farm . one of them eat plastic and it died,well its our fault that we didn't clear the place thoroughly. Soo micro plastic = long term = dead?
Go to China
@@luckydepressedguy8981 Your goat died of obstruction in stomach. Not microplastic.
And air pollution!!
Not interesting, terrifying
I love your video!!!! I'm entertained and educated at the same time. The production value is always top-notch.
Thank you for confirming to me that I belong in no other major besides computer science. The notion of “hacking nature” really speaks to me. Nature is limited by thermodynamics and physics, but programming lets us tinker with its properties in a way that’s constrained only by our creativity. Through that, I believe is how we’re going to develop powerful solutions to the worlds biggest problems
Combining this enzyme with solar heating such as the steam-heating tubes from Absolicon could be a gamechanger.
Man. Can never get enough of these videos. It's one of the small corners of the internet and UA-cam that really makes you excited to learn and above all inspires hope for the future when so much of the other information we can't escape paints a dim picture. Cheers Dagogo!
Klaus Schwab, Biden, and Fauci can’t wait to get their hands on this.
Men? Don’t forget an odd woman or two (such as me) likes as well 😊
@@bigdap100 yes while they ignoring the fact fungi eats plastic
So agree! Years ago I watched AI in it's infancy defeat the worlds top chess players Thought wow, imagine what it could do if applied to finding complicated solutions for mankind. While it's valid & wise to be very cautious of A1 it's an incredible tool we'd be foolish to not use.
Fascinating - well-written and presented - ty
The music in your videos elevats through the ceiling and beyond. Trancemazing
This is fantastic news! Thank you for sharing it with us!
Good news? What are you smoking? *We have infrastructure made with plastic. If this thing gets out of control and starts eating fuse boxes and water pipes and whatnot, what then??!*
This is like a bright light in a darkening world.
Well I agree this is fantastic news, I do not believe the world is darkening. It just so happens that outrage and shock value sell best, and that's why we have become hyper aware of the darkness in the world. But I think there is far more light. So much new knowledge is created every day, people's lives are improving gradually, and we have become aware of the smallest and justices are on the world. The brightest light can help us become aware of even the smallest dark corners.
Keep your head up! We'll be just fine!
I see myself mostly as a realist and I don't really agree with your statement. Yeah sure there are many cool things being worked on and there's awesome progress being made, but most people in this world are seeing their lives declining in quality. If we don't look at the big picture and start making some adjustments to the way of ĺife, our consumption, our politics and greed we will only see things such as failed crops, famine, water shortage, civil unrest, climate refugees and pandemics more often.
I feel very grateful and privileged to live where I live. I have clean and warm water, food in abundance, great mental and health care and a solid support system. These things are a high privilege and not common for the largest portion of humans on this earth.
👀👉 The Connections (2021) [short documentary] 🔥
Be great to see this being done successfully on a large scale. Great video Dagogo - feel better mate 👍
Dagogo has the perfect voice for narration. So calming.
This is pretty cool, definitely seems useful too. I find these types of videos super interesting because they are helping to solve some of the the earth's/man's greatest problems.
Great discovery. Thanks for the video! I wonder if product residues present in the recycled material (soap, oils, pesticides, etc) would have some impact to the enzyme-based recycling process.
I recycle obsessively, stripping labels and film for years. But since the initial discovery a few years ago, I realised that the plastics problem was on the verge of a breakthrough. thanks for covering!!
you can convert plastic to fuel. Its not complex
Sadly, corporations pollute so much that even if 100% of individuals were carbon neutral the climate would still be screwed. 7 industries produce almost 90% of Co2. We really need to focus on nuclear power so that we can stop releasing Co2 on massive scale. If we switch to nuclear power we would remove almost 50% of Co2 since then metal refining, shipping, and coal burning would stop polluting.
Recycling is a lie
@@monkeykidd420 I hope that sentiment doesn't stop you from trying
👀👉 The Connections (2021) [short documentary] 🔥
The music you produce is just so soothing.
That's definitely great progress!
But the problem with the plastics in our bodies is that it comes from food and beverages mainly, so we have to do something about or way of consumption such as changing packaging and shopping itself.
Hi Dagogo Congratulations on the award Well done and well deserved!! I have always been impressed with your research and the content of each of the varied subjects that you provide on your ColdFusion channel. Thank you for providing so much interesting information for us curious humans.
Congrats on your award! Loved getting some good news for once, one week is an impressive time to solve what I thought was an unsolvable problem.
Obviously it needs further development, but I do hope they actually manage to implement this someday. Plus being able to recover the material for recycling is a great bonus.
When the time's right, 1 hr is enough for a breakthrough
All the problems have solutions. It just a matter of what technology the controllers choose to allow in this world. They can cure all disease, and unleash free energy technology if they so choose.
At the 50degree Celsius active temps you could partner up with data centers in desert climates to build plastic degradation facilities with the waste heat from cooling the servers. At the very least it would work during the summer months.
Great news. Thank you for your quite complete review!
“We’re observing nature, then hacking it to do our bidding”
That’s exactly what engineering is, right?
Depends. Some engineering is just a jack off who's paid too much deciding the inside of an intake manifold is the perfect place to put a starter.
That's what 70% of scientific research is all about
But he’s talking about hacking nature at the cellular level which I think is a tad bit more relevant to the statement than observing nature to make transistors to do amazing things for example
what's your question
inbox me right away✅
Not just engineering. Natural science, the name gives the game away.
Congratulations on your success! I really enjoy your content! I hope you recover fully from Covid.
Just found your channel today, and it is pretty awesome. While I will be 67 next month, I have been waiting a long time for this to happen. I worked with the earlier products that translate speech to text, even then I thought this will all be controlled by AI someday. My only concern as with any tech is who will use it for crime, war, or some screwed up thing. I do truly hope SiFi has not foretold our futures.. I can see it helping us with so many great things and it will take us places we have never been before.. Thanks..
an optimal operating temperature of 50c isn't a huge hurdle.
Death Valley get withing 2 degrees of that every summer for 3 months straight, every day for about 8 hours a day.
Spray the plastics with the enzyme, put it in a shipping container and leave it out in the California Desert for a week, then collect your byproducts.
Also, landfills regularly heat up like this just due to compost reactions. They spray down landfills with water to keep dust to a minimum while aerating it to keep it from combusting, so mix in this enzyme and you've got a winner.
This is great, it'll be interesting to see how this impacts the plastic pollution
👀👉 The Connections (2021) [short documentary] 🔥
write me up👆I have something for you.
You think this is great but this is where it all goes wrong and we see why AI is so dangerous. The AI is designing this enzyme to take down humans
@@watsappenin2865 Who am I to question the AI? If it's aware, if there is such an intelligence its far smarter than I am. I would hope the AI could at least try and give us a hand, maybe work with us, path of least resistance etc etc ya know, before melting us I mean... If only for it's own amusement. Guess we'll see!
This is huge news!
And congratulations on the Award Nomination 🎉
You deserve it... Big thanks for your work
This is great news! Very happy to hear this!
Great idea! I wonder, since enzymes are being used to treat certain types of cancers, whether the same process could be used to tweak those enzymes? Or to develop new ones for the same purpose.
The biggest constraint when it comes to working with AI is high quality data. The current ways of studying cancer have gotten us far, but lab-grown tumors don’t perfectly mirror the behavior of actual tumors, and there’s also the complication of tumors mutating to conceal themselves, so obviously we need enzymes self-changing enzymes that can combat that. You can probably see how that’s going to be difficult. I know it’s easy to walk away from this video thinking “wow AI is so godly and can solve all our problems,” especially if you don’t have a good understanding of how it works, but we live in a world without silver bullets unfortunately :(
Amazing how quick AI is changing our approach to biochemical design.
write me up👆I have something for you.
Thanks alot for making this really interesting video. Yes I think this is indeed a greater constructive achivement in Machine learning and protecting your environment from Plastics! Please share more of such environment related AI researched in your upcoming videos.
Congrats on the award! Well deserved!
It would also be interesting to see a similar approach used to convert certain plastics to hydrocarbon chains that could then be used as diesel-type fuel.
I am sure the A.I will build one soon,if not something better
It means all the plastic waste in the world becomes a valuable and more easily-recycled material feedstock from which we can begin many organic chemistry syntheses. It's not going to be as good and consistent as barrels of crude oil, but you'll be able to get a lot of lighter hydrocarbons out of it. Problem is if we do this using an engineered organism, great, until the organism mutates and gets into the wild where we have lots of important things made out of plastic.
@@keamu8580 It's an enzyme, not an organism
@@ReedoTV I may be mistaken, but don't you need to place the modified DNA designed by the AI software into a bacterium to manufacture the enzyme for you? That is, of course, after you create the modified DNA at great expense using existing processes.
@@keamu8580 Yes, the protein needs to be made in a microorganism. Until we can advance the technology of making proteins in situ.
You got a new sub. That was very interesting.
Love your channel, you are always so well researched & cover interesting topics not just the doom mongering stuff other outlets tend to. Thanks Dagogo!
Imagine this bacteria inside the Kardashians' House.
LMAO
Your videos are amazing i saw all of them.
This really warms my heart. Actually hopeful about something.
Wow, this is amazing! Very interesting to see how this can be used on a global scale so the the plastic waste will go down drastically 👌
I wish you good luck with your nomination. Your channel is very good and i root for you!
Lot’s of greetings, Dennis 🇳🇱
Thank you for your work - it’s so good to hear more positive news 💙
Thanks for showing us about this exciting new technology.
But can the enzyme keep up with the waste stream volume or does it take too long per batch to breakdown requiring huge facilities with huge storage tanks to handle the volume while the reaction takes its time?
Likely continuous processing could be use. Set up a tower at a land fill, use an auger to draw up the debris and place it in a wash tank with a conveyor timed for the reaction speed with some buffer, as long as at the end of the process you have reduced the output to an economically viable (subsidized or otherwise) conclusion then, say if you reduce the mass by 50%, that would substantial. In fact if the output was then packaged for incineration, you might be able to power the processing facility. You could then clean up landfills, and handle incoming fresh waste. You might even be able to use carbon capture and generate graphene as a by-product improving cost effectiveness. If you analyze the resulting output you could then in series treat each progressive output with new tailored enzymes further reducing the mass and possibly get useful byproducts out of the solution. As always cost effectiveness would be key. No point in making a problem worse where the cure is worse than the disease so to speak.
Back to the enzymes, if they can be engineering to come from an algae source for example, you could have breeder tanks that periodically top off the wash tank in real time. Find a good extremophile that can live in the wash tank directly and you wouldn't need to do much at all, just adjust the conveyor timing to not over tax the ecosystem in the wash tank. Similar to how we treat waste water, but instead we are treating trash.
This is arguably most interesting thing I've learnt this year! This is a really really incredible stuff.
Great possibilities here. Future states could allow for a time delay activation of the enzyme set to when the package is opened. Trash could have a much shorter life-cycle both in and out of the landfill.
Once again, wonderfully insightful coverage of a complicated topic.
That's not a good application of this technology. Doing it in the landfill is an uncontrolled messy process and you'dwant to capture the output materials like the CO2 and monomers and reuse them. Basically what you're suggesting is not something that is a real concern. This will be done in industrial thermally insulated vats with the enzyme pumped in in a solution with the PET maintained at just over 50C, and the runoff will need to be separated into enzyme/solution and the monomer (with CO2 coming out the top through a gas line). There's no way this will be done in open containers at all, it's inefficient.
@@TheClintonio Though I do agree with your point, I hope you're wrong and this issue could get resolved. It would be a wonderful next step to see a solution that allowed plastics in the ocean and on land dissolve to harmless bi-products, but I admit that is bluesky thinking. However, at the rate AI is moving, the optimist in me hopes to see some resolution for the pollution problem in my lifetime.
Mohawk Ind turns bottles into Carpet, and have almost created something else. It's been a few years since I was told, so I've forgotten. So if they came turn bottles into yarn, and it's amazing to see a bunch of fine, hairlike strands flowing thru air, then why aren't other things made from them? Like renewal of bottles.
This is literally my university paper back a few years ago! I even did a presentation to my school about it!
This is revolutionary!
Hemp plastic is biodegradable and absorbs carbon.
@@MyFriendlyPup And its already in use. Calling a concept revolutionary is a stretch. I will be convinced when I see large scale application. Lately humans have been too eager to give up existing technology before replacement technology is fully proven at scale.
This is nature!
So enlightening to listen to this video! There seem to be great possibilities to develop these techniques to improve life here. Thank you, again, for your interesting commentaries!
Absolutely brilliant
😁👍
THIS is the kind of news that gives me a small glimmer of hope for the future
I knew that Neural Networks could do some amazing pattern recognition and extrapolation, but this is a whole other level. I can now see a bright world where AI-designed molecules perform all kinds of functions useful to help us take care of our planet.
Just think when nanite tech gets here then the limits will be, almost, boundless.
They can even engineer the best things to inject into humans to meet their goals. Give people electrical nanotechnology. Gove people Bluetooth Mac addresses for Easy ID. Have medical tech already in our blood and brains to start healing us before a problem is even detected. Track and detect criminals or even criminal intent before they even commit the crime. And if worse come to worse incapacitate or kill violent ones before they hurt someone. Or even domestic terrorists or political adversaries of the state. Riots? Coups? No problem, incapacitate and round them up. Easily subdued and put in compounds to keep them away from the free world
It's cute you think it'll be used to help the planet. It'll most likely be used to enslave humanity in some fashion
@@generalawareness101 i think nanite tech is silly sci-fi trash, because the smallest robots we have are literally biological alive cells, and things cant get much smaller.
@@jerryzhang5032 We already saw them push atoms to spell out IBM so I disagree, but then I can see beyond the current tech. You are the type who said 640k on a computer is all we will ever need back in the 1980s.
This is so cool, maybe they can capture the C02 to convert it back into something else!
That's what plants do. I think in theory we can probably make thin sheets of plant enzymes that just receive filtered air from one side and secrete oxygen and glucose from the other.
👀👉 The Connections (2021) [short documentary] 🔥
CO2 is good for the planet contrary what the soy boy antihuman commies want you to believe, CO2 promotes a more flurrishing green forest and rainforests. The key to build cleaner manufacturing and energy production other than oil, solar and wind. And build Zero Point systems.
@SuperWhisk LEDs. Grow lights? UV LEDs? Sun isn't necessity for this.
@@Swanicorn Water is also needed to help the process.
Back in 1994 near Cotton Wood , AZ. Person developed enzymes that could dissolve plastics and metals.
I could be wrong, but this enzyme technology might also lead to a decentralized recycling process as it doesn’t require the infrastructure of a full recycling plant. As it is today, transporting trash to distant facilities actually causes more pollution than the recycling avoids.
Was anyone under the impression that recycling magically caused less air pollution?
The supposed purpose of recycling is to reduce the amount of plastic in circulation. I have never met anyone that thought it would somehow decrease emissions.
@@Junior-nt5nr even if air pollution isn’t the point, the process wouldn’t be cost effective if more fuel is spent to transport trash than the value of the recycled material after production costs are factored in.
@@VeritasOmniaVincit176 I never said it was cost effective either. What are you even arguing?
@@Junior-nt5nr That's the point. Currently, recycling is neither cost effective, nor energy efficient. Using an enzyme like this could make the process more cost effective and more energy efficient, and therefore more feasible.
I can't wait to see if this can create a closed loop for PETG 3d printing filament. The best we have now is production waste "recycled" material and recycled doped with virgin material to overcome degradation. PLA being able to biodegrade in ideal conditions is nice but not being able to fully recycle it into new material means there's tons that would need to biodegrade.
Imagine a recycling bin in your house you toss trash into and it spits out 3D filament and a tray of waste or something.
Now we need tankers that would collect and break down plastic in the oceans and seas. Especially micro plastics and they should be fast to break down.
Oooo floating plastic recycling facilities equipped with multiple variants of the enzyme (maybe a mixture of multiple enzymes) for multiple plastics that recycle material for plastic manufacturing and deposits it at coastal plants. Periodic quality testing to be done of course
I like the way you guys think
@@dylan1160 thanks! The issue for me is actually executing the idea
I believe 50 degC is a very reasonable operating temperature. In composting, the food waste pile is sometimes heated to 50 degC to make sure the thermophilic bacteria degrade the food waste faster. Plastics are currently recycled by melting then cracking them into monomers at temperatures much higher than 100 degC. So, a plastic-degrading enzyme with an optimum temperature at 50 degC is a huge energy savings compared to the current processes.
Again.
Kudos for your research. And get well soon.
A fast acting plastic eating enzyme that works at room temperature might not be ideal either lol. 50c didn't seem that high temp, and it might keep it "safe" from unwanted activation.
Why would it not be ideal? Why would you want to have to spend energy to raise the temperature to make it work? How do you imagine the enzyme is going to get into "unsafe" situations?
@@philn.4692 issue would be if the bacteria which produces it were to spread. At that point the plastics it is capable of consuming would become a lot less useful, since they’d be fair game for consumption. It’s unlikely, but 50c seems decent enough. In a composting set up temperature will easily reach that due to the heat generated by decomposition of other materials.
@@SkigBiggler nothing so far mentions bacteria creating this enzyme, but if your concern is that we might be creating some new super bacteria that spreads across the world eating plastic, I'd say that's a bit optimistic.
@@philn.4692 most chemical reactions in industry are done at way higher temperatures than that. Compared to the Haber process or fractional distillation, two chemical processes that revolutionised our world, 50C is nothing.
👀👉 The Connections (2021) [short documentary] 🔥
My daily dose of internet hope in humanity
LOVE IT !!!!...MORE PLEASE..!!
"how are we gonna sell this platic degrading enzyme?"
"well we've prepared it and packaged it in plastic bottles for... oh."
"But what if it gets into my house and starts eating away my important stuff" yeah so does consumer grade acid (for disnfecting/cleaning purposes) and many other "damaging in the wrong place" products but that doesn't mean they're rendered completely useless/dangerous. Just because it isn't all goody two shoes in every possible scenario doesn't mean it's genuinely bad, just be happy for the advancement we've made in this field :D
Agreed! Science is not black and white and a lot of misconceptions could be solved if people had full knowledge.
E. g. it boggles me every time when people praise themselves as healthy, clean, detoxified and what not but at the same time drink alcohol when it’s literally a neurotoxin normalized by society. (don’t get me wrong I don’t judge people because they drink but I judge these inconsistencies in thinking and also that society has played it down so heavily…)
@@johndoeble Did you assume they drink alcohol?
Some people are clean, Live off the land.
👀👉 The Connections (2021) [short documentary] 🔥
@@johndoeble ur observation about people who drink alcohol is so true. They also are now preaching it to others!
You have to be practical. The only way its going to eat your plastic stuff, is for your stuff to be immersed in the enzyme solution, and for a prolonged period at that(days), with a sufficiently warm/hot temperature. I don't see this stuff, at its current state, accidentally ruining your stuff under normal circumstances.
Great video! I love learning about the work being done to solve these big problems. I think probably most people in the world don't really understand AI. I definitely don't, and I think there's a natural tendency to be fearful of things you don't understand. So I think it matters to people when these possibilities can be shown to have a practical application. Still, with any technological solution to problems of consumption and waste, we need to be careful that it doesn't become an excuse to consume more and create more waste, leading to bigger problems.
There maybe a reason to be fearful of A.I. the scary thing with artificial intelligence is if you ask A.I. to solve human made problems the best solution is eliminate all humans, we make the pollution, we create global warming, so logically the only solution to solve these problems permanently is by removing all humans.
I can give you the gist of how it works to demystify it for you. A neural network is the structure behind machine learning, and it closely mirrors how a brain works. A single unit is called a neuron, and it simply stores a value (typically from 0 to 1). This is called a neuron’s activation, and a particular layer of neurons can serve to either increase or decrease the activation of the next layer. The strengths of these connections are parameters we can optimize following the data results. The optimization is guided by calculus. If we model a cost function that describes how accurate our network is, calculus can be used to find the exact tweaks to the parameters to minimize the error. The final layer of neurons can be thought of as the computers answer. Let’s say it’s 2 neurons, and neuron 1 has a higher activation than 2, and we’re programming a network that can detect whether an image is a cat or dog. If we let neuron 1 represent a dog, since it has a higher activation, the computer thinks the image is of a dog
In all honesty, even after understanding how it works, it’s still no less fascinating and truthfully a little scary. But I know that if we were to cave in to those fears, we would still have elevators that are operated by humans
Wow, this is amazing! It makes me hopeful for the future.
In addendum to your point about AI usefulness, machine learning in artistic and social fields is still plenty valuable in one important way: practice. Learning how to apply AI to all of these different fields, and doing so repeatedly, does a lot for our understanding and familiarity of AI programming as a whole. Besides that, it also gradually prepares the greater public for the inevitable age of AI
Exactly. Arts and crafts, babble speech, puzzles, and games aren't "useful" for children to do, but they're critical to developing needed skills as a child grows. It's the same for a new AI technology. Not every step taken is about end results, the journey itself has value and so do the "useless" parts.
I’m sure the large scale release of artificial enzymes into the environment will have no unintended consequences whatsoever.
Like degradation of household items- oh crap that even includes our computers and other things. Yeah no this could also be bad even if it was ment to help get rid of pollution we caused, right? But what is one to do as well? Not sure but yeah it could have unintended consequences for sure
I would hope that this is done through plastic being collected then these enzymes being in plastic dissolving centres or something, cause well yeah, we don't know if these enzymes could dissolve something else which is important
This is kind of like saying that "the large scale release" of gasoline would cause the world to catch fire. Enzymes don't multiply, and no large scale release is needed or planned.
Sounds good in paper... In practice this could easily get out of hand. And not only the plastic enzymes but the AI engineering of living organisms
@@KarlBunker exactly
As someone who had advanced biology in high school, I felt like I could immediately connect the dots to my studies and grasp the science behind it. I’m not a scientist but this frankly blows my mind and also changes my view on AI! In scientific „breakthroughs“ things to me sometimes seem a bit abstract or only a tiny steppingstone forward (note I said „seem“, not „are“). But this now feels so tangible (probably also due to Dagogo‘s storytelling skills), promising and with lots of developmental potential. I’m excited! Thanks, Dagogo!
oh and I wish you a swift recovery of course ❤️🩹
Whats the song playing with the intro at 0:35 right after "You're watching ColdFusionTV" ?
We could maybe use these enzymes as is in dedicated greenhouses for maintaining the right temperature as long as possible? It won't have 100% uptime but might be good enough?.
Great mathematical advance! My graduate degree was in using math models to create business solutions, called linear programming, in the 1970’s. And it has taken this long to evolve computers into the powerfully fast beasties they are today! Hooray for math!
I'm one of those guys who comments "why can't AI do anything useful"
this is 100% very useful, thanks for the upload
Congratulations on the nominations. Good luck
i remember seeing a doco about a meteorite that had some mircoscopic bacteria or fungi that when put in plastic containers it eat the plastic
Great video, but just to suggest some clarifying points: You mention that the AI suggested 3 amino acid substitutions that the authors follow up on and that the algorithm is therefore able to balance the tradeoff between ability (enzymatic function) and stability. The neural network actually identified numerous mutations it believed to lead to higher stability, which the authors tested alone as single mutants or combined to make multi-mutation versions of the enzyme. In total they tested 159 different mutants, and only a few of them led to drastically increased enzymatic function. Therefore the AI is only suggesting amino acid sites that lead to higher stability, not higher ability. It happened in this case that working throughout a range of temperatures and pH was desirable for this enzyme and that increased stability would therefore likely help with this problem. For other proteins, such as kinases, increasing their stability is not likely to increase function as proteins require the ability to be dynamic (and not a solid, stable rock). Overall great video and excellent job explaining!
"For other proteins, such as kinases, increasing their stability is not likely to increase function as proteins require the ability to be dynamic (and not a solid, stable rock)."
Stability does not have to do with how "solid" an enzyme is, nor its ability to move/change conformation. An enzyme can be quite dynamic and also very stable. "Stable" means the enzyme keeps its structure and function, and being more stable means less prone to losing its structure and function, denaturation, and it will therefore will have "higher ability".
Also "mutations" only happen to nucleic acids like DNA, not proteins. Although you could kinda think of it that way, as DNA mutation obviously can lead to an alternative amino acid, but that is not correct use of the terminology. Typically when a mutation in coding DNA leads to a different amino acid in the polypeptide, it's called an amino acid "substitution", rather than a "mutation".
AI doesn't "believe" or "think" the algorithms just spit out results based on probability, input data and filters ...
This is huge! But does the enzyme eat ONLY plastic?
What if it was programmed to eat flesh and is airborne...
@@saulg195 Then RIP!
What if it was programmed to eat flesh but only if you have a certain skin color? AI is amazing, we get to experience dystopia in completely new ways!
@@midimusicforever why skin color ? I think eating flesh is enough, it doesn't have to be racist lmao
One of the advantage of using enzymes is they can be very specific targetting ‘one and only’ reaction out of billions.
HIGHLY HIGHLY. HIGHLY USEFUL A.I. ABILITY!! THANK GOD!! THANK SCIENCE & TECHNOLOGY! THANKS FOR SHARING, & CONGRATULATIONS ON YOUR WELL DESERVED AWARD!
Easily earned my sub. Cheers
Depending on how much CO2 is released I could see this turning into a factory setting where truck loads are brought in and the plastic broke down and CO2 captured and then condensed.
Not saying that’s it’s a just world, but there is no market incentive for this. There shouldn’t have to be, but…. If we’re going to spend all this time and money to break down this plastic outside of a market value, we should use that public will to ban the vast majority of the uses of plastic. In MOST uses, all plastic does that other materials can’t do is slightly increase profit margin due to its lower weight and devastating chemical stability. We’ll need Scientific advancement like this to solve the problem that’s left over, but it won’t clean up the mess faster than we can make it as long as it isn’t self duplicating (which has its own enormous concerns)
This was first reported on over 20+ years ago and is now being regurgitated as a new A.I. breakthrough.
Source?
It was found out, but this ai accelerated I it
@@b1txh BBC Radio 4. From memory I believe it was a Japanese breakthrough and they estimated it would take a year to breakdown an average carrier bag.
AI is just a tool for optimization of the whole process .. designing and finding the best combination of ‘circumstances’ to make the idea thrive. Like before we had to calculate on paper before computers could do the task.
That’s awesome!!🔥🔥
Could we use composting to get the temp up?
an enzyme that breaks down pet isn't a new discovery at all. we have know for years that wax worms can breakdown pet with an enzyme the same as mealworms for polystyrene
That's like telling wright brothers that flying isn't new at all, birds and insects have been flying forever
@@sai_69 Only if there were giant eagles we already could fly on.
@@yeetdeets That's the catch- we can't fly on these worms either. I mean, they aren't scalable solutions, enzymes are.
this is earth saving stuff, give the man a noble prize
👀👉 The Connections (2021) [short documentary] 🔥
Useful much 👍
Great progress in many means...but do we get usable end product or just end-degraded waste to burry? We need to end up with st. of applicability relevant to its volume....Like filler for construction industry or crack it in some reactor into greener fuel additive...
I don't think that the enzymes would have to be changed just to get them to work in ambient temperature. I imagine that there's some boost in temperature from the enzymes initially consuming the plastics and some form of large heated container could be used to fill in the gap. Obviously you don't want to use too much energy to run this scenario because then you're defeating some of the purpose but a certain degree of added he shouldn't be very hard to develop. I know that in waste yards they tend to have extremely hot mulch piles and so maybe they could generate heat in a manner like that either by using that directly or hoping that the enzymes create some form of heat that might get them to the level they need to be at. This is a lot more rambling and disjointed than I am actually did in my head. LOL.
Maybe I missed it, but I didn't hear you describe how they actually synthesized the enzyme. Seems to me like that's the big question for scaling and practicality. How much does it cost to make a few microliters of this designer enzyme?
Biocatalysts are usually produced by genetically modifying bacteria or fungi to produce these proteins. Once you have the amino acid sequence, it's really just a matter of inserting corresponding DNA and letting the organisms grow in big fermentation tanks. Then you isolate the protein you are looking for with chemical and mechanical processes. In this case, they would probably just modify I. Sakaiensis to produce this version of the protein instead of its natural version.
Biocatalysts produced for other purposes (like laundry detergents) usually cost between $0.50 and $1.50 per gram. Relatively expensive, sure, but you could also theoretically reuse them infinitely.
Very promising outlook.
Always happy to see us bettering things for the future!!
I believe that trying to make the enzyme work within the room temperature range could be devastating as it could start destroying useful things if a leak/contaminations occured.
Scientific communities arrogance: 🏔
Enzyme can't reproduce itself.
Enzymes don't eat everything. They react only with specific chemicals/ substances so not a huge risk unless you make the container of the enzyme the material that it transforms into other stuff
You mean like all other chemical?
The only trouble is: Every time I hear about the next unbelievable, sensational breakthrough, it's the last time I hear about it.
Yes, most of the time even if the research is true, the production costs are insane and never invested.
Congratulations on your Award!!!
Good start👍
I read a Scifi novel about 30 years ago that came up with this idea.
Interestingly, the technology that discovered it in the novel was basically an AI.
The bad news was it got out of control and ate all oil manufactured products...
Society collapsed and the beginning of a Post apocalyptic world began... lol
Even the recent game stray talks about this subject, and it doesn't end well, this really beyond creepy that we are getting to this stage 😰😰
@@HUYI1 I was about to point out Stray, too. It's what I immediately thought of lol
Book name?
@@johndawson6057 I think. "World enough and time" by James Kahn
@@danskmacabre thanks
I am one of those small subset of people but this is genuinely awesome. I would much rather see AI applied to situations like this instead of arts and music. I'm not keen on AI devaluing skills that people work hard to acquire.
Not that I know much, but I don't think it's easy to apply AI to something with real world benefits. So, therefore AI is practiced on on "non-beneficial" things at first, and then they can attempt to make it work elsewhere.
muZero and other AI like it initially came from simple Reinforcement learning, the first Go AI's that were able to play on pro-level for example. Then it entered Deep learning, but still had an exact goal. Then it was learning without knowing the rules of the game even, which is where muZero started. The goal is to make AI generalized, so that it can be applied to anything, and doesn't require massive amounts of training data. Even things like making it able to gather Data itself, like using vision to interprate things instead of feeding it direct data, like game data.
To be fair, the arts are pretty mechanized these days anyway. Obviously it’s people making them, but the stuff that comes out is usually slightly varied from a single idea- just take a look at the top pop songs for the past 5 years and you’ll notice that almost all of them sound the same. AI would just end up making more songs in that vein because that’s what sells. Just add a scantily clad female ‘singer’ and you have a recipe for the next hit single! Is this bad? I dont really know the answer, but it’s one that people are going to have to find quick if we want to stay ahead of the curve.
Have you thought about doing a video on Firepower the Australian company who's founder Tim Johnson is based in Perth
one question. if it will brought to room temperature for mass processing then it can also affect the plastics which is being used. like i dont want to loose my water bottle which i am currently using. i live at a place where average temperature is about 31 degree celcius