WB retweeted
AI in trading is a race along a speed–accuracy Pareto frontier, and firms should try to push that frontier outward at every point.
While you’re thinking, others are trading, so a crude deterministic computation can capture edge before prices adjust as long as it is merely directionally correct AND it is the FASTEST.
Trades which come after (based on the same data) are adversely selected in that they happen only after someone who is faster has already traded a price to what they think is fair.
Slower inference can still be useful but has to earn its latency in that it needs to find additional INFORMATION based on the same DATA to compensate for the opportunities that disappeared during that time.
The race to be first to market with new information happens at every level of inference complexity.
In principle, each point on that Pareto frontier offers a potential edge in which marginal additional intelligence and model complexity contains information the market has yet to priced in.
That suggests firms should invest across the entire speed-accuracy frontier with:
(1) Extremely fast deterministic engines for immediate reactions. (2) Fast local inference for decisions that require some interpretation. (3) Large models on well-connected datacenter clusters for deeper reasoning.
These are complementary capabilities where the fast system acts on the first-order signal while slower system looks for missed implications or mistaken interpretations.
1/ A large hedge fund asked me in to talk about open and local models. The conversation made me realize institutional capital allocation will rapidly change over the next year.
They're following the frontier closely, but wanted to discuss how open and local will affect how capital gets distributed across AI spend.
WB retweeted
in 2 years every company will have 2-5 guys who get paid $10m/year to be the only ones allowed to use fable 10
your job is to figure out how to be one of those 2-5 guys
WB retweeted
The presumption that the Fed raising short-term rates reduces inflation is predicated on the belief that higher rates reduce demand and investment.
But what if higher rates don’t reduce demand and investment because the demand for intelligence and energy is unaffected by higher rates because winning the race for super intelligence has a near infinite ROI and the demand for compute will remain incalculable.
Why won’t higher rates at this unique moment in history therefore lead to more inflation as interest costs are embedded in everything?
And the problem is compounded as the more the Fed raises rates, the more inflation we will have and the more the Fed will need to raise rates further and so on.
But what if the old models don’t apply to the current paradigm and the Fed is wrong?
I think the Fed might have just made a mistake. Am I right or am I wrong?
WB retweeted
ADAPTIVENESS MATTERS
The conventional wisdom in investing is that you can't time the market, shouldn't try, and will only hurt yourself if you do. Parts of the industry have elevated this into a cult of passive: whatever happens, just keep buying.
There is some logic here. Market action is mostly noise. People tend to overweight irrelevant datapoints, inappropriately extrapolate recent events, and pay high transaction cost drag when tunrover is too high. Trading too much destroys capital.
But “people often respond badly to useless information” is NOT a justification for “you should never respond to useful information.”
Ignoring all incoming information when deciding what to own is clearly crazy. For example, if you really took the industry's advice to keep it passive to heart, you would have been buying bonds with negative nominal yields for years in Europe. This wasn't difficult to understand at the time. There are clearly certain datapoints that should matter enough to inflect investment behaviour. And the world is changing faster than ever before - you've got to be able to react.
I view adaptiveness as a design choice with trade-offs to navigate, just like any other design choice. How much and how frequently you change your positions should depend on:
1) How materially the information changes your assessment of future returns and risk.
2) What it costs to change the portfolio relative to the expected benefit.
Reacting appropriately to new info does not require having an opinion on every headline.
Of course, the hard part is establishing that the information you are paying attention to actually helps. Making adjustments is a lot easier with LIQUID, capital-efficient instruments (like futures) in your toolkit. And clearly "less is more," in the sense that under-reacting will tend to draw your results closer to a still-acceptable passive-like outcome, while too-much overreacting will nearly guarantee permanent capital destruction.
Still, I think you can do far better than passive alone, by paying attention to the basics, like: risk conditions, capital efficiency, awareness of macro factors, momentum, carry returns, and the like.
Finally: the industry's advice to "keep it passive" really amounts to "you aren't smart enough to take even the obvious stuff into account, and not only that, we aren't either."
That's wrong and it really shouldn't take alot to do better. The key is a disciplined methodology for judiciously turning evidence into positions, with enough humility and restraint to mostly leave the noise alone.
WB retweeted
was showing my normie friends codex and how you can just make whatever you want and they literally didn’t give a fuck
WB retweeted
Ok Astra is also terrible at writing research papers. I asked it to generate novel ideas about optimal portfolio diversification and it came up with a number of trivial or overblown observations, written in an unmistakably annoying AI style.
Fable was marginally worse.
Actual creativity is still far, far away.














