Claude Makes Bio Discovery in Wet Lab, OpenAI Solves 100 Math Problems

0:00Welcome to the podcast today. We want to talk about the one of the newest breakthroughs that Anthropic has had in biology. There's a bunch of drama, of course, with Anthropic and Dario saying we need to pace the frontier. And the same time they were saying they were opening a wet lab in San Francisco after 21 hours of the wet lab being open is already made its first discovery and Dario has been talking about it, which has led to a lot of excitement, but also backlash from people concerned that Anthropic is going to slow down the development of a lot of great scientific things. There's people worried that they're going to create the next COVID. So there's all sorts of sides of the conversation. We're going to be breaking down what's going on and what they're discovering in biology. As I think this is one of our most advanced movements when you have a company like Anthropic putting the full force of their research power behind AI and science. This is going to get interesting today on the show. Before we talk about that, I wanted to mention if you use Claude and you use something like Claude Coerck, which is an amazing tool for getting stuff done on your computer and you're frustrated that Claude Coerck doesn't have the

1:00ability to generate images, audio or video. I would love for you to check out my own startup, which is called AIBox. We created a connector that lets all of the other AI tools be generated inside of Claude. So I recently had an experience where I wanted to get a ton of images generated for a project I was working on and I had this connector inside of Claude and I told it, you know, the types of images I was wanting, but they were all open AI images to be generated. Claude was able to go generate over 600 images and put them all into the project and the traffic on that site. Consequently, it is now skyrocketing and it's up like 10x in the last three months. So overall, there's so many incredible capabilities. Claude doesn't have them all. If you want to add all of the audio image and video generation capabilities into the projects that you're working on with Claude or with any other AI tool like Muse or ChatGbT work, you can get the AI box MCP connector. There's a link in the description. It's only $8.99 to get started. So try to keep this affordable for everyone. Link in the description. Get all the AI models to work together.

2:01Okay, let's talk about the big announcement that Dario made. It was in response to Anthropic coming out with a tweet. This is just two hours ago where they said Claude has discovered a previously unknown enzyme system hidden in the DNA of bacterial graphs based on the enzymes gene are beside the enzymes gene sits a long array of repeated DNA structure that looks somewhat similar to CRISPR. Okay, so Dario came on and gave kind of a breakdown of what's actually happening over on x. He said that they were announcing the Claude lead discovery. So obviously it's not completely just Claude running all by itself. Claude lead meaning like Anthropic is funding this. There are real scientists involved. But what they're doing is having a bunch of Claude agents run a bunch of experiments to try to figure stuff out. He said that he said we suspect this could represent a new gene editing mechanism. It's precise function biotechnological utility if any or level of significance is not yet clear. So you know they discovered something new. They're like look we found something new that no one really knew about. We don't know if this is important.

3:04If it means anything this came after I think 21 hours of their wet lab being online. So it's not like you know there's a lot of research that still needs to happen. But at a minimum I mean this is really exciting to have sped up. Dario said the work was done mostly. They're not entirely by Claude. Our life sciences team suggested a broad area of research. Claude read through the literature and a bunch of genome data and discovered something interesting. Then Claude proposed experiments to verify the discovery and our team carried them out. Okay, so right there I think a lot of people are concerned. And by the way there are actual companies working on like basically getting robots attached AI models and the robots are mixing the chemicals and the vials of doing all the experiments themselves. And some people are freaked out about that. In this case there's a research team and they're like hey there's probably some things to discover in this area. Perhaps those researchers have already done research in those areas and know that there's a lot going on there. So they point cloud of the right direction. Go look at these things. I mean it's just like you or I going into a

4:04domain that we understand deeply and we're asking cloud or chat to help us with and do some research so we can figure stuff out. It's the same concept. So like it's it's quote-a-quote Claude lead, Claude's the brain. But like a lot of these models are doing the research for everything nowadays. But whatever I digress I do think it's cool. Anyways, Claude goes does some research says hey look I think I might have found some things go go do these experiments and they have humans going and doing the experiments. Okay, Dario said it's easy to dismiss this as a one off or curiosity. But we've repeatedly seen a pattern where AI performance in our in new intellectual domains goes from week to superhuman in a matter of a few years in 2023 models struggled to do math at the level of an average high school student in 2024. They started to do well at math competitions for the best high schoolers in the country. In 2025 they started to solve minor open problems in early 2026 more significant open problems and in late 2026 they're beginning to solve the top few open problems in all of mathematics. We believe AI for biology is on a similar

5:04exponential trend. What they're specifically referring to here is that we actually have a you know a big kind of a big kind of thing that came out from an open AI where they've actually put together like a math advisory board they're calling it at IAR at IAS after they had open AI apparently resolved about 100 open problems. So you know open AI there's a lot of drama when they solve a big famous math problem that other mathematicians have been studying are working on and people you know speculated that they were getting people's chat conversations and open AI to NIS. Regardless a lot of mathematicians right now are some of them are saying hey you know AI is ruining the the field of math you know we can't intellectually work on stuff I think there was nine mathematicians that were you know recently there's a whole I think 25 field metalist winners signed an open letter earlier this month they were warning that AI labs are quote threatening mathematicians intellectual work because AI could do a lot of the work so there's obviously drama here but in response to all

6:09of this open AI has formed kind of a math advisor group where they're like hey look guys we are getting a lot of math done we're solving a lot of problems we would love to put them back to actual mathematicians to go verify and check what's going on this group that they have in particular they put together they said like the group isn't allowed to tell open AI to slow down or they're not allowed to do as much research it's just an advisory group that's going to help oversee the checking of a lot of these things so apparently opening AI says look you know there's one big famous one we have a hundred other math problems that are big famous math problems that we've solved and a lot of these like there's you know million dollar prizes so a lot of mathematicians are excited and working on this and some people are you know complaining like I mentioned 25 people signed that open letter really like look you're you're ruining mathematicians ability to do like complex stuff but at the end of the day these AI tools are just calculators and so just because they figured out how to do something and they're a really complex calculator in my opinion that just means mathematicians should be able to solve more problems and if these are problems that have been around for a hundred years unsolved gee wouldn't it be awesome to just solve them and move on with our life like I don't really care if there's a quote unquote field of mathematics where it's people's job to solve these problems

7:13and figure and spend a lifetime figuring out trying to win a million dollar prize if we found the answer to a problem we didn't know the answer to just publish it and get it over with now implemented into whatever it's useful for and let's move on with society so that's my opinion and I'm sure yeah triggers people or whatever one thing that I do think is interesting is Darios obviously very aware of this whole you know situation is happening in a math and he's using this as you know kind of a reference point saying look the same thing is going to happen with biology Dario goes on in his tweet to say the main difference between biology and mathematics of course is that math can be done purely theoretically while biology requires experimentation some of you's this to draw the conclusion that AI's utility and biology will be limited we think this is wrong as we've demonstrated today humans can collaborate with AI to perform the experiments validate key results in a few weeks and if necessary work with AI to iterate on what they find eventually it may even be possible for cloud itself to safely perform the experiments by autonomously controlling lab equipment with appropriate safe guards in place but we aren't doing that today our lab is also a BSL one BSL 2 facility that doesn't

8:17handle materials dangerous to humans okay I think he's really trying to show like a the possibility and the you know opportunities that are come out of this but also you know try to hold back so he's like look like this is advancing really quickly it's going to be amazing there's lots of cool things we can do cloud could eventually perform all of these like I was talking with autonomously controlled lab equipment he's like but don't worry we're not doing any of that today because obviously he just freaked out around a couple weeks ago saying that AI you know could end the world with bio weapons and then he's like you know got this got this wet lab going Dario then goes on to say more broadly biomedical advancement has many stages from fundamental biology discoveries to translation research to drug discovery clinical trials and finally the actual delivery of medicines and health care to patients we are also interested in these later stages but even simply accelerating the first stage of fundamental biological discoveries has the potential to speed up and broaden the entire pipeline improving our understanding of biology and sharpening biologist tools can

9:17drive forward all of the later stages for example by identifying new drug targets finding new therapeutic modalities allowing for more precise measurements and speeding up the experimentation or experimental loop which itself further accelerates our understanding of biology this will this will not in itself speed up clinical trial times but if this but if it succeeds it could greatly increase the number of promising candidates that go into the pipeline and increase in throughput even though latency remains basically he's saying look like just because we're in the discovery phase of a lot of the stuff and people are like well you should go through clinical trials and I could take years he's like yeah but if we can get you know 90% certain that a drug is going to work today instead of having to pitch 90 different drugs and do all the experiments if you're at which one actually works we are speeding it up because we're getting more high quality drugs tested faster I mean it's a good like he's basically saying there's a huge pipeline to getting drugs into into people's hands like you know life saving drugs but even if we're just speeding up the beginning of the pipeline the whole pipeline does speed up and he says eventually they will go downstream and try to work on other elements of

10:22the pipeline and speed up the whole thing which yeah there's all sorts of opinions on but you know at the end of the day if we can get actual drug discoveries actual medical discoveries actual breakthroughs in biology that are helpful for humanity why wouldn't we want to accelerate that to the absolute utmost I'm sure everybody listening has a family member or no someone that has you know suffered with a disease a sickness cancer all sorts of horrible things if we can fix and cure some of those I would want that done as soon as possible personally for for my family okay he says in machines of this Dario Ken and his tweets you know very long tweet but he said in the machines of loving grace which is a blog post he wrote he said I wrote about AI AI's potential to cure most diseases in five to ten years by the way I'm gonna stop for a second here I just love how like Dario quoting his blog post it's like he's quoting like you know like a famous book everyone showed no about like in the count of Monte Cristo is like in the machines of loving grace anyways you gotta know that it's a blog post he said a goal that sounded impossible right

11:25curing disease in five to ten years but one I believe is just barely possible if AI is applied to every stage of the pipeline the first step is showing that AI can first help with and then drive biological discoveries so this is what they're working on this kind of first step of the pipeline he said Claude's discovery is the latest in a line of related prior work that goes back decades beginning with systems like CRISPR and continue with with discoveries like the bridge recombinase and VIPR in the past few years he then says recently there has been heightened interest in systems based on reverse transcriptase RT enzymes and enzyme underlying and the enzyme underlying the system clot identified and most recently a Stanford team working independently described a novel RT system with an associated non-coding array that is in some ways similar to the one Claude found though they are distinct systems that evolved independently from each other I believe that we are at the very beginning of finding such systems and developing them into

12:27powerful tools for biotechnology so the one thing that I will say here there was a group over at Stanford that that essentially published a paper on how they found like a different enzyme not the same one you know they evolved differently but you can see like you can basically backtrack how Claude did this right they looked at the Stanford researchers team looked at what they did in a very narrow area and said okay if we applied that discovery to everything what's left to discover and I don't want to say low hanging fruit I mean you know this is something that's been running for a long time but that's kind of the concept here and I don't think that's a bad thing that's not derogatory to Claude at all I'm sure some people I bet you will those say like look there's you know Claude's just ripping off of other people's work I mean when you're discovering drugs the safe people's lives I don't care who's ripping off of whose work let's just get the drugs discovered and I don't really care who gets the credit either if I'm being honest but that's just me like I just want you know people's lives to be saved and improved but it is interesting right like people have made discovered discoveries in narrow areas and sometimes for funding reasons not or for or for you know just the they just don't have enough manpower there's a lot of different

13:30reasons why they stay narrow on those discoveries they don't apply it to the entire field and so if there's a discovery in a very small narrow area AI can grab that and run experiments across a ton of different fields and you know kind of bride in the horizon the horizons of of what was found and you know it's it's more worthwhile so from you know there's more discoveries to be found so I think that they will probably see a lot of this and people will criticize it and say like hey look AI is not able to actually discover anything on its own yada yada and maybe there's elements of truth to that maybe it will improve over time like Garyo says math has done at the end of the day though I'm really not concerned if if AI truly can go take other people's discoveries apply them to the entire field and find new areas that we haven't seen I don't know why we don't want it to discover as much as humanly possible not humanly possible as much as machinely possible okay finishing up the the tweet from Dario he said I'm proud of the resources and Throboc has invested in accelerating the public benefits of AI through the life science and we're aiming both to grow our

14:30life science team and to work with other scientists to extend this approach to a broad range of problems if you have a proposal for a research collaboration or interested in joining our life sciences team please reach out I mean this is probably the best place for him to put a plug like hey guys look like if you want to join the team if you have any other good discoveries that you're on the verge of making we would love to to help you with it I think it's two sides of the sword which I again do not do not do not give a crap about to the to the smallest degree one is if you're working on something cool like the Stanford team wise and you're close to a discover you have a discover you think you figured something out and you go to Dario like hey can you fund this because they're not just gonna you know fund budget to anyone like they're funding the budget they're giving you the lab they're working on it they're putting you in a round and amazing team they're giving you some great tools they're giving you resources to a lot of anthropics kind of tools in a way I would just say like it's the same things if you went to a big pharmaceutical company with a you know perhaps a discovery and they were funding it like you're finding a funder now the downside that I think some people say is like well you did all the work to figure that out and now anthropics most definitely going to do a big press release like this and take a credit to which I say who cares if it's

15:34something that's going to help humanity if Dario wants to take credit for it I mean you'd be great if you could get your name on there for getting it started but like let's just get the discoveries out there for humanity and if he's funding it and you're helping and working on it like you should be excited about it but whatever at the end of the day guys this is really exciting there's a lot of incredible things being discovered a lot of drugs I know I've been pessimistic and I think a lot of the internet if you read the comments on there it was like no you know opening eyes or anthropic is gonna release the next COVID we can we can now create COVID 10 times faster with anthropic and stuff I think honestly like if we're being serious anthropic is a is full of a good team their life sciences team are good people they're trying to make good discoveries I don't think there's anything to Ferris going on and at the end of the day really these are real humans I mean at this point maybe that'll improve and progress in the future but right now this is still just real human teams pointing the AI model and anthropics resources in a direction I believe they can find something once it finds something it's a real human team going and doing the experiments so overall I mean I think there's a ton of human in the loop a ton of oversight a ton of making sure that this

16:38is done correctly and safe anthropic knows how bad you know it would be if anything were to go wrong so I think we're gonna see some you're gonna see top level safety and I think we're gonna see some really exciting discoveries coming out of all of this guys thanks so much for tuning into the podcast if you haven't ever left a review on this show it would help me out so much to get a review from you so if you appreciate it if you learned anything new I would love to get a review on the show from you drop a five star review and drop a comment I read all of them so I really do appreciate them let me know if there's any other topics or companies or things that I should be covering on the show and I'll catch you all in the next episode