Earlier this week, the Latest York Times shone a light-weight on among the desperation that founders are experiencing as they struggle and fail to secure compute power for his or her nascent artificial intelligence startups, due to the big firms (and even wealthy nations) racing to grab them up. One founder reportedly said of the graphics processing units, or GPUs, that he needs for his company: “I take into consideration [them] as a rare earth metal at this point.”

Based on that Times piece, founders try quite a few measures to amass the chips, including calling in favors from friends at large equipment vendors that might need GPUs to spare, and navigating an obscure U.S. government program called Access.

At the least one firm, the worldwide investor Index Ventures, happened on an extra idea, it told the outlet. To assist ensure its portfolio firms aren’t hamstrung by the shortage, it struck a cope with Oracle to supply its founders with a few of these sought-after chips (specifically Nvidia’s H100 chips and Nvidia’s A100 chips).

To learn more concerning the arrangement — one which other enterprise firms are undoubtedly trying to duplicate —  we talked earlier today with Erin Price-Wright, a Bay Area-based partner with Index who focuses on enterprise software and AI and who, before joining the enterprise firm in 2019, was the top of product for Palantir’s data analytics and machine learning platform. Excerpts from our chat have been frivolously edited for length and clarity below; you possibly can hear our longer conversation here.

TechCrunch: Tell us about this partnership with Oracle. 

Erin Price-Wright: Access to compute is certainly one of the largest challenges that AI firms face, and it’s especially hard for an early-stage company to get their hands on GPUs. It’s less about the fee specifically however the incontrovertible fact that something like greater than 95% of GPU capability is already allocated to large players on this space [because] they make these pretty big pre-commitments with cloud vendors. So in case you’re an early-stage company, and also you’re just attempting to start training, or tremendous tuning the model, there’s often a extremely long lead time between when GPUs are even available. It will possibly be three months to a yr in some cases and it’s really hard to only start.

For those who’re an early-stage company that’s still determining what your product is, you don’t even understand how many GPUs you wish. So even that means of discovery of understanding what your workloads are going to seem like may be super difficult for early-stage firms. So we’re partnering with Oracle to supply GPUs to our earliest-stage portfolio firms, because we wish to assist remove that barrier of access in order that they will really give attention to what matters from day zero. Ultimately, the goal is to assist all of those firms graduate to their very own cluster. We’re not within the business of providing these massive GPU clusters to our firms. . .but we really need to provide them a head start, in order that they will start constructing faster as a approach to help level the playing field.

How did the deal come together?

We desired to be certain that people who find themselves constructing against very tangible business problems didn’t feel like they’d to vary their business model or change the way in which they were representing themselves or change the way in which they were fundraising in an effort to just get access to GPUs. So it was really born out of seeing this pattern repeatedly with early stage firms where we were like, ‘That is where Index as a fund actually has real leverage. And we are able to use our position available in the market, our relationships, and the incontrovertible fact that we are able to sort of aggregate this demand across multiple firms to essentially provide value-additive services’ [to our founders].

Did Index put a down payment together or has it purchased chips outright from Oracle? Are you giving Oracle a stake in these startups?

We’re not purchasing any chips outright. So the partnership with Oracle is that Index makes the pre commitment on the behalf of our startups and pays the cloud bill. Oracle manages the cluster — they’ve been a unbelievable partner — after which our firms get access to that GPU cluster free of charge.

So that you’re paying [this cloud bill] upfront. Did you’ve gotten to speak together with your own investors about that? That’s not typical of what [a venture firm] would do historically.

When it comes to the actual structure of how the agreement works, I’ll probably hold off on sharing too a lot of those details.

Is that this an exclusive relationship? Is there anything to stop other enterprise firms from doing the identical thing?

Yeah, in fact [they could do the same], there definitely isn’t [an exclusive relationship with Index].

One profit that Oracle gets out of it’s to fulfill the subsequent generation of unbelievable firms as early as possible. Within the means of using our GPU cluster, we’re actively helping our firms navigate the means of signing their very own dedicated cloud deal. So the concept shouldn’t be for them to [do] this in perpetuity; it’s for them to develop relationships with Oracle and AWS and the opposite large cloud providers and sign their very own dedicated contract.

One in all your portfolio firms, Cohere, counts Oracle as certainly one of its backers together with Nvidia, that are two of the businesses you most need to have involved together with your portfolio firms at once.

One in all the ways we actually may also help our portfolio firms is ensuring they’re connected to the best people at the best time, in order that they get the resources they need.

Index has at the very least 20 portfolio firms that fall into the AI/ML bucket, including Cohere [which has already raised $445 million] and one other company that recently raised an enormous seed round, Mistral AI in France. Is simply too much money being invested broadly in generative AI or are we still within the ‘early innings,’ as VCs prefer to say?

We’re within the early innings. I do think we’re rapidly entering a cooling off period by way of sentiment, especially for a few of these very large rounds and particularly from traditional VCs. There’s still a extremely big gap between the promise and power of the core models of technology and what it’s going to take for them to be actually used and useful across many use cases within the enterprise. There’s just an enormous infrastructure gap missing that should be filled, and it’s not going to be filled overnight; it’s going to take a while.

Over the approaching 12 months, while I’m still very excited concerning the power of the core technology and the way transformational it’s going to be for the world, I feel we’re going to see slightly little bit of a backing off as firms really grapple with it, determine the ROI, sort of prioritize use cases and begin actually constructing real things beyond possibly the one or two prototype demo apps that they’ve been working on for the last six months. That’s after we’re going to begin seeing the infrastructure emerge that’s going to begin supporting these use cases at scale.

How do you as an investor be sure that your AI firms don’t overlap? And is that any harder or harder than in relation to traditional startups?

I don’t think it’s massively different than how we take into consideration competition elsewhere. Everyone paints AI as this standalone category. But when I look forward even two years, let alone five or 10, each piece of software that we use may have AI as its beating heart. There can be no piece of code, no software, no application, no website that you just visit, that doesn’t have AI as a core component of it. I almost give it some thought like SaaS. Is each SaaS company the identical? No. Each SaaS company has a database, each SaaS company has a front end, each SaaS company has some interaction between the 2. AI is sort of much like a database in that respect. It’s just sort of a core constructing block in the way you construct software.

We’re very early available in the market, so there’s going to be some movement and a few change as firms determine easy methods to use these tools and what specific problems to go after. But it surely’s not different than how we take into consideration traditional SaaS investing from my perspective.