'Not a big part of the work': Meta's LLM bet has yet to touch its core ads business

"We are not by and large using LLM architecture to do ranking and recommendations work yet. LLMs are not a big part of the work in core ranking and recommendations today. Eventually, the hope is that one day they will be because LLMs don't just optimize what already works, they can reason about content and context in ways that current systems fundamentally cannot."

"Today's ranking engines are built on engagement signals - likes shares and watch time - and those signals require scale, and scale requires time. It's a feedback loop that works extraordinarily well at Meta's size but it has a ceiling. It can only optimize for what users have already responded to, making it inherently backward looking and blind to content or context it hasn't encountered before."

"LLMs can bypass it entirely, reasoning in real time about whether a piece of content is likely to interest a specific user based on what the system already knows about them without needing the engagement history to learn from first. That's the capability Meta wants to bring to its core products."

Meta's ranking and recommendation system, which powers content and advertising decisions across Facebook and Instagram, currently does not meaningfully utilize large language models despite their potential advantages. The company's CFO Susan Li acknowledged this at a technology conference, noting that LLMs are not a significant part of core ranking work today. Current systems rely on engagement signals like likes, shares, and watch time, which require scale and historical data to optimize effectively. However, LLMs offer a fundamentally different capability: they can reason about content and context in real time without requiring extensive engagement history, potentially overcoming the backward-looking limitations of existing algorithms. Meta views LLM integration as a future opportunity to enhance its core products.

Read at Digiday