Google just ran a three layer strategy move in one day at their cloud next conference. They announced new TPU chips. Chrome is turning into an AI coworker and a multi-billion dollar compute deal with Miriam Radi, the former co-founder of OpenAI, her company, Thinking Machine Labs. I think this is one of the clearest singles from me so far that Google is structurally perhaps ahead of OpenAI and Amazon in the AI stack.
Before that, though, I want to talk about the fact that OpenAI is teaming up with Infosys to get Chate GPT into 60 plus countries of enterprise deals, Bloomberg reports, an unauthorized group breached anthropics new cyber tool, Mycos, and there is a new research lab called Neil Cognition that has landed $4 million to build agents that actually specialize like humans. So we're going to get into all of that on the show today. Our first story is that 10x science. This is a Stanford spinout of a Nobel laureate, Carolyn Bertosi's lab.
They just closed a $4.8 million seed, which was led by initialized capital. What they're doing that's so fascinating to me is that basically there's this problem where models like DeepMind is protein predictor. They're spitting out thousands of drug candidates. So there's just thousands and thousands of these drug candidates.
And there's a huge bottleneck in pharma where it's not just about getting all of these candidates, but it's actually triaging on it. It's actually figuring out which of all the candidates is worth pursuing to try to make medicine or therapeutics. And basically, the standard triage tool is just mass spectrometry. So it's very slow.
It's very hard to interpret. It's domain experts only. 10x science is basically just building a sass layer on top of that. And they have deterministic chemistry plus AI agents to try and make the analysis traceable and explainable, which basically matters because regulators don't accept a black box answer on what the molecule does.
So just because an AI model is like, oh my gosh, we discovered this super cool thing. This isn't going to fly. Regulators don't want that. There's no, we can't have the black box on molecules.
So we have to be able to fully understand it, fully test it. And knowing which of these chemicals, which of these drugs to test is a big problem. So everyone in the AI biotech conversation is basically talking about the generative side. And almost nobody is building the kind of picks and shovels layer that's underneath of it.
And so this is why I think 10x science is an interesting company. The next thing I want to talk about is Neo-Cog Mission. So this is an AI research lab that came out of stealth with about $40 million in their seed round. You sue an Ohio State professor who runs an AI agent lab there is the founder.
And the round was led by a Cambium Capital and Walden Catalyst Ventures, Intel CEO, LitBoutang, and also Databricks co-founder Ion Stocca. Both wrote angel checks into this. So I think that's a really strong signal, right? When you have the CEO of Intel, and when you have these high profile co-founders of something like Databricks, this is phenomenal.
But essentially the thesis on this company that I think is important is that the current AI agents succeed maybe 50% of the time because they're basically unreliable generalists. And so what they're arguing is that humans aren't great at doing tasks just because we know everything. We're great because we specialize fast when we're dropped into a new domain. So Neo-Cog Mission is trying to build agents that self specialize the same way instead of the current model where you hand craft a custom agent for every vertical.
I've built enough custom agent workflows to know that kind of this per vertical approach doesn't really scale. You run out of engineers before you run out of use cases. So if Neo-Cog Mission can actually ship an agent that learns the rules of a new environment, and if it's doing this on its own, which I think is definitely gonna be the key, I think that is a massive win. Right now we have only 15 people on their team.
It's mostly PhDs. Definitely an early company, but I think this is one that's worth watching. Okay, so we have some bad news for Anthropic today. There is a report from Bloomberg yesterday that an authorized group got access to Mipo, which is Anthropic's new exclusive enterprise, cyber security AI tool that's gonna take over the world, they gave it to a handful of enterprises because it's so dangerous and good at finding security vulnerabilities.
I think this is probably not good news because they've been a lot of hype about how dangerous it was and they only give it to special people. And if it really is getting leaked, that's not great for them. I think what basically happened if you read the reporting of Bloomberg, there was a group that operated out of a private Discord channel focused on Unreleased AI models and they got credentials from a third party contractor working with Anthropic. They then guessed the models endpoint URL based on the pattern anthropologists for other models.
And that's it. They then gave Bloomberg screenshots and a live demo to prove access. Anthropic's response all this is that they're investigating and so far they say they found no evidence that it touched Anthropic's own systems, just the vendor environment. I think this is interesting because this is kind of where a lot of these security incidents are happening right now.
It's not really a model exploit. It's just a contractor credential and plus kind of a predictable URL pattern. And my advice that I would give to people is that if you are a company that is running AI tools, make sure that you are auditing who is on your vendor side and who has access to all those tools right now. And kind of the bigger thing is that this is also happening in other places.
Anthropic is actively in early talks with Goldman Sachs, JP Morgan, Morgan Stanley about an October IPO. So any sort of bad PR story for Anthropic right now is not gonna be great. They definitely want to get on top of this. Next up I have some OpenAI news, OpenAI and info Sys just announced a deal to push chat GPT and specifically codecs or coding tools, kind of the competitor cloud code into Infosys enterprise client base.
This is across more than 60 countries. They don't really disclose the terms or how much money is changing hands in this. But TechCrunch did report on this earlier today. And basically for context on Infosys, they did about $267 million in AI service revenue last quarter, about 55% of their total.
And they play in kind of the massive enterprise transformation deals where Microsoft and Accenture often take the system integrator slot. So right now, Infosys is getting OpenAI models baked into their Topaz AI platform. And OpenAI gets a channel into Fortune tier accounts that they don't reach directly. I think OpenAI is trying to make a big push for enterprise right now.
I mean, obviously this has been a go for a long time, but this big push is coming right now as it feels like Anthropic is really running away with the market in enterprise. I've worked with a lot of enterprises that bought the chat GPT enterprise licenses and they kind of sat on them for like a year because they didn't wire them into anything real. This is basically the gap that I think Infosys is trying to fill. And strategically, I think this is kind of mirroring what Microsoft already does by bundling a zero OpenAI with Copilot except Infosys plays in a lot of deals where Microsoft doesn't always win.
It's not in the deals. It doesn't have the distribution. So it's kind of a new channel. I would expect to see a deal like this to help push OpenAI's revenue higher.
OpenAI obviously wants enterprise revenue to catch up to Inforopic, which is now over $30 billion in annualized revenue. They need this type of distribution in order to try to catch up to Inforopic. Okay, if you're still paying for chat GPT, Cloud Gemini and Grock, by the way, 11 lives for audio maybe like separate image generators, I have to tell you about AI box. This is what I personally have built.
It's what I recommend to my friends who keep asking me how to actually use AI without going, you know, without spending a ton of money on subscriptions. Basically, AI box gives you access to over 80 AI models in one place, all of the top models. So you can pick what everyone is best for the task that you're working on or just ask a question and we'll pick the best model for it. Part that I think is super useful is our automation builder.
You could just describe a tool that you want in plain English and it will build a workflow for you. So you don't need to code, you don't need to wrestle with a new platform, you just describe it and it will build. It's $8.99 a month, very cheap. I think this saves you a ton of money on just all the different subscription platforms and you get your hands on every single AI model that is worth using.
We have a ton of news from Google, their Google Cloud Next conference kicked off today and Google did not come empty-handed. They had three major announcements, new TPUs, Chrome is an AI coworker and I think in machine labs compute deal. And I think taken together, basically they're gonna make Google the most structurally positioned player in the AI stack right now. I'm gonna walk through all of these different deals but yeah, I've been pretty impressed by their announcements and of course we can see other players but I would definitely say never count Google out in this industry.
So first the chips deal, Google announced two new TPUs. They have the TPU-8T for training and the TPU-8I for inference. The split alone is I think what's really interesting to me. Inference is the dominant cost of running AI and perraction right now and having these kind of dedicated inference silicon matters economically.
It makes a big impact to the bottom line. So Google's claims is that it is three times faster at training and 80% better at performance per dollar against the NVIDIA alternatives and the ability to scale more than a million TPUs in a single cluster. So they're hitting a bunch of different things here but NVIDIA has traditionally, I don't know this necessarily even in NVIDIA's fault although I'm sure some people will disagree with me here but NVIDIA traditionally has had a problem with the cluster sizing and I remember when opening I was not opening I when XAI did their first giga-term and whatever, they have a robot named for it so now I can't remember but they put together a hundred GPU, 200,000 NVIDIA GPUs. It was like this big deal they kind of figured this out and other people started building spigger and bigger clusters.
Well, and it was kind of like they had to figure out a lot of infrastructure and build it and think about it. Well, it looks like Google is building straight for that and the ability to have more than a million TPUs in a single cluster is insane and I mean insanely scalable. So I think those are Google's own numbers. We don't have any independent benchmarks yet.
Google's also not killing the NVIDIA partnership. They're reselling Vera Rubin in Google Cloud later this year. So I think basically Google wants both chips and it wants to let customers choose which is kind of the opposite what AWS is all in one training. But is right now, I think Google will actually come out ahead on that.
The other thing that I think is interesting that Google announced is that they're turning Chrome into an AI co-worker. So they have a new feature which is called auto browse, Gemini powers it and it's basically running inside of your Chrome for workspace and it reads context across your open tabs and automates actual workplace tasks. So it can enter your CRM data, it can compare vendor quotes, it can summarize candidate portfolios, it can write up competitor research. If I'm being honest, it sounds like an interesting tool.
It sounds useful and I've actually tried some similar things. Microsoft had something like this in Microsoft Edge a while back where all of your tabs that are open, it can just get data and read and interact with them, which is super useful. Sometimes though, you need things that are not in your tabs and I think this is where Cloud's Co-work is still gonna keep crushing because they're app, you can access to your desktop, it can get files, you can download things on your local device. So I know that like, Google isn't trying to directly compete with that.
They're kind of saying, they're just like this human in the loop and they're not letting agents execute without your approval. And so anyways, I think they're kind of pitching as a plus in my personal opinion, it's behind the ball, it's kind of not as good as something like Cloud Co-work, which actually sometimes I don't want to have to get 100 pop ups asking for approval. I just wanted to get the job done, especially when I'm seeing it do it like three, four times successfully, I don't want to approve it, I want it to automate. And so maybe they're playing a different target audience there, but I still would probably use Cloud Co-work over that.
You can also, something that is very useful though, and kind of like Cloud Co-work, you can save frequent workflows as skills and you can fire them with a forward slash. This is really not in the US, workspace users getting this first. And there's also a commitment that enterprise prompts are not gonna be used to train Google's models, which I think is the right call. A lot of enterprises would be very concerned about that, especially if they wanted to put confidential data in there.
But again, and not to like, I don't know, to say all of the downsides to this. Again, though, you have to go and fire off a skill manually, where something like Cloud Co-work, you can just set these things up to automatically run. Okay, the next thing that I thought was interesting is that Google Cloud has just signed a multi-billion dollar deal with Miriam Raddi Sync Machine Labs. This is a single digit billions it gives thinking machine labs access to NVIDIA's GB300 system on Google Cloud, plus training and deployment services for the first product, which is called Tinker, which is a tool for building custom frontier models.
Thinking Machines is raising at a $12 billion valuation and they run a heavy reinforcement learning workload, which is basically very insanely compute intensive. Mile-aut, one of the founding researchers has basically talked a ton about how much Google got them up and running really quickly. So I think this is something that's going to be helpful for the company. The thing that I think ties all these together is that Google is running a three layer strategy, bottom layer silicon, TPU plus these resold NVIDIA's.
And then they don't really care which chip wins. I mean, obviously they kind of want their own to win, but they're going to make money either way if you're going to be buying video from them. The middle layer is being the compute host for Frontier Labs. And Thoropik already runs on TPUs, and now thinking Machine Labs is going to be running on Google Cloud.
The top layer is the agent layer in the browser in the workspace. Google is the only player that's hitting all three credibly right now. And Thoropik is obviously crushing it with end users and kind of in that top layer, but they're not making their own chips. And so Google has a big advantage when they own the entire stack.
Microsoft is really heavy on opening eye at layer two, but really weak on silicon. Amazon is betting the farm on Thoropik. NVIDIA is a chip layer and doesn't touch the application layer. Google is basically the only company really doing the full stack.
So I think the strongest argument against this is that Gemini still hasn't closed the gap with GPT 5.4 or claw on opus 4.7 on the benchmarks that consumers actually care about. And so I think maybe those TPU numbers are Google's own. And you could also say, until an independent lab publishes real world training runs on the AT and on the AI, the three X claim that they've given is marketing. And I mean, I don't want to doubt Google too much, but you have to be fair.
There is a lot of these AI companies in Google's been known to do this in the past to make big claims on marketing. And it doesn't always stack up in the real world. But here's why I still think even with all of that, Google is in an incredible position right now. The customer really doesn't care whether Gemini tops the evil leaderboards.
If they can run inference cheaper on Google stack, and if they can serve it to their employees through Chrome and wire it through the workspace data without having this big, huge system integrator, structural position, I think matters way more than just pure benchmark lead in the next 12 months. And Google integrated with every tool that people are using right now. And so I think Google's going to do quite well. The other thing that I think is interesting is that Google is doing this with a DOJ antitrust kind of suit overhanging them right now.
The search remedies case that they've been in for a while now is still live. And turning Chrome into an agent layer that pulls from workspace data across your browser is basically the exact kind of move that makes regulators pay attention. I think you would be wise to watch for some comments coming from the DOJ or the EU in the next few weeks. This is going to draw scrutiny that Microsoft and their opening ideal has already been dealing with for a year.
What I'm watching personally is the independent TPU-80 and AI benchmarks, whether auto browse graduates from pilots to real rollouts inside of enterprises, and whether more frontier labs migrate to Google Cloud, now that thinking machine is in and anthropic is already there. So I think if even one more lab follows, that is a huge structural shift. All right. That's it for the show today.
If you've got something out of this, it would mean the world to me if you could drop a comment on Apple Podcast or if you could hit the stars over on Spotify. It's the About tab on Spotify where you can leave a review, leave some stars, drop a comment on Apple. It helps the show a ton, right? I don't ask for it for nothing.
It really helps. More people find it. It helps in the algorithm. Also, if you want to access over 80 models with an automation builder that runs in plain English for $8.99 a month, go check out aibox.ai.
There's a link in the description. In the case, I'll catch you guys all in the next episode.