Email or username:

Password:

Forgot your password?
jonny (good kind)'s posts Post Back to profile
jonny (good kind)

The #LLMs aren't just weird text generators, and when these companies talk to investors they don't talk about whether they're sentient or not. They talk about "understanding intent" as a synonym for matching search queries to ads. They're parsing your email and calendar and docs and matching them to entities in their knowledge graph to predict your likelihood of clicking an ad. They don't talk about generated text as thought, it's to optimize ad content and give better clickthrough rates to advertisers who pay to embed in the answers of "LLM-type experiences"
abc.xyz/investor/static/pdf/20

First, Google Al. It's important to recognize that our advertising business has obviously benefited over the past decade from the transition to mobile. More recently, we had outsized growth in advertising revenues during the pandemic, with 2022 advertising revenues $90 billion higher than in 2019. Going forward, we're focused on growing revenues on top of this higher base through Al-driven innovation. Sundar highlighted the incredible opportunities underway with Al and the transformative impact it will have on businesses. Already breakthroughs in everything from natural language understanding to generative Al are fueling our ability to deliver results that drive meaningful performance for advertisers and are useful to users Take Smart Bidding, which uses Al to predict future ad conversions and their value, helping businesses stay agile and responsive to rapid shifts in demand.
bidding performance, allowing us to move advertiser outcomes down the funnel to drive better ROI and use budgets more efficiently. In search-query matching, large language models like MUM match advertiser offers to user queries. This understanding of human intent of language combined with advances in bidding prediction, are why businesses can see an average of 35% more conversions when they upgrade exact match keywords to broad match in campaigns that use a target CPA Google Al also underlies our Creative products, like text suggestions in Google Ads and creative optimization in responsive search ads. We're excited to start testing our Automatically Created Assets Beta, which uses Al to generate headlines and descriptions for Search creatives seamlessly once advertisers opt in
We have already deployed - if you look at the impact of things like BERT and MUM have had on search quality, making search multimodal, driving the usage of products like Google Lens, feel like we've been scaling up well. In Google Cloud today with Vertex Al, we've already been bringing Al APIs to Enterprises, and they are on a pretty healthy growth path. So we do see secular opportunities ahead, both in terms of putting these APls out, making sure every developer, every organization in the world can use it. And as I said earlier, we're in very, very early days. And I think there's a lot of room ahead. In terms of Search, too, now that we can integrate more direct LLM-type experiences in Search, I think it will help us expand and serve new types of use cases, generative use cases. And so I think I see this as a chance to rethink and reimagine and drive Search to solve more use cases for our users as well. So, again, early days. You will see us be bold, put things out, get feedback, and iterate and
4 Apr 2023 at 7:00 | Open on neuromatch.social
2 comments
jonny (good kind)

I'm not saying LLMs are magic and can do all the things they promise to investors, I'm saying these companies don't care about whether the bots can think. they won't work and that's worse: what they certainly will do is deepen the logic of surveillance that drives their application in advertising and provide a lot of flimsy, bias ridden, nonfunctional LLMs as platforms to data consumers like governments, cops, and insurance companies to make use of surveillance data under the cloak of LLM datawashing.

4 Apr 2023 at 8:03 | Open on neuromatch.social
jonny (good kind)

for the one billionth time, anti-capitalism is the excluded viewpoint in academia.

why are we talking about how to detect students cheating and not why our neoliberal academies bankrupt our pedagogy by turning our classrooms into grade mills.

why are we talking about whether the #LLMs can think and not how they exploit the ideology of convenience they have trained us with so every time we want to know something the energy of a small bomb is expended in generating an answer, computing an ad profile, and nurturing dependence to choke off independent access to information at scale.

for the one billionth time, anti-capitalism is the excluded viewpoint in academia.

why are we talking about how to detect students cheating and not why our neoliberal academies bankrupt our pedagogy by turning our classrooms into grade mills.

why are we talking about whether the #LLMs can think and not how they exploit the ideology of convenience they have trained us with so every time we want to know something the energy of a small bomb is expended in generating an answer, computing an ad profile,...

4 Apr 2023 at 8:18 | Open on neuromatch.social
Go Up