Ravi Matra

understand the technology deeply and think about the trend line for where technology is going, not where it is.

Investing in technology is all about almost thinking a little bit outside of yourself to understand that we're on a exponential curve and not just plodding ahead linearly.

And that forces you

to have to understand deeply what technology might be capable of and imagining what's possible and then getting a sense of being right on the timing for when that knee in the curve and inflection is going to happen and arriving either just before or at the time when that inflection is starting to emerge.

time for our morning brew daily davos update with lightspeed we are joined by lightspeed co-founder and partner ravi matra ravi we've spoken to a number of founders in the lightspeed portfolio when you're looking at a startup to invest in what's the fastest way you separate signal from noise we look for founders who think about what we call the super cycle or the trend line and not the state of the technology today ai

intelligence is evolving and improving so rapidly, and that relates to some of the underlying exponential trends in the drop of cost of compute.

Inference cost has gone down by something like 100x in the last year.

The performance of AI computing, the ability

at one third of the cost this year versus last year to produce the same level of intelligence.

These things really mean that founders have to be focused on where the intelligence technology is going and not where it is today in the context of their idea.

And if that doesn't compile in a way that we think is robust, we will not proceed.

Two things can be true at once.

If you look at the underlying substance, it is a larger investment in a technology shift since electrification.

We had just in the last year four companies investing almost $400 billion in the compute infrastructure AI build-out.

Moody's is predicting there will be over $3 trillion invested by 2030.

And that's across compute, data centers, chips, power.

So that being said, there's probably some amount of funding that's going in to companies where there's a circularness to what they purchase and what they sell in the AI domain.

AI is not really a market.

It is a paradigm shift that is very horizontal and intelligence touches so many industries.

We think the transformation and the value is very real.