It's easy to say what tech companies want from AI, but much harder to talk about the product strategy - they're all pretty much the same. "Just build a better model!" Where does that go? Can they differentiate? What would it mean to differentiate a product that can do 'everything' - what would different everythings be?
For the past decade, Benedict has given an annual presentation on the state of technology, and he did the latest at Slush in Helsinki last month. In this episode we discuss some of the challenges and issues that he tried to cover. You can find the full presentation here - https://www.ben-evans.com/presentations
Benedict went from being a consultant to an analyst to having his own business, but in essence, he's always been an analyst. Toni went from policy to consultancy, then from B2B sales to Formula 1. After four years of doing the podcast, we thought it would be interesting to sit down and discuss how we got here and talk all things Formula 1.
A quarter century after 'don't be evil', a judge has found that Google is abusing its monopoly in search. But no-one knows what happens next, and whether this ruling will change anything. Will Apple build a search engine? Will ChatGPT change search? Does it matter? There are many more questions than answers.
As we go into the summer, we know a lot more about generative AI than we did six or nine months ago - or at least, we have better questions.
Generative AI is the thing, and all new software will be built around it. But while everyone is experimenting and some people are getting huge value out of ChatGPT or Midjourney right now, others haven't worked out how to make it useful. Yet. So how do we find use-cases for a universal, general purpose, magical technology, and is that a crazy question?
Will the US finally break up Tiktok? Will the EU break up the App Store? And why does Temu want to keep your orders under $800?
Are there questions that an AI chatbot shouldn't answer? Should it always give the 'right answer'? Are you sure? Google has egg on its face this week, but this isn't easy, and with generative AI, we're going to re-run all the arguments and panics we had over content moderation in the last decade.
We’re past peak TV, the charts are curving down, and Hollywood is pretty sure that streaming was a bad idea. On the other hand, music is growing strongly and might even end up bigger than CDs. Why have newspapers, books, movies, TV and music coped so differently with the internet?
Yes, we bought one. What’s it like and what can we say that we didn’t say last summer? What has Apple built, what is it for, what does it mean for Meta, and why does it cost $3,500?
Everyone needs an AI strategy (there was an email from the CEO!) but what would that mean? How does a big company work out how to deploy a new technology? How is this the same as every other platform shift, and how might it be different?
Every year, Benedict produces a big presentation exploring macro and strategic trends in the tech industry. Here are some of the key takeaways from this year's presentation - AI, and everything else. Presentation - https://www.ben-evans.com/presentations
We spent the last 30 years building structures on top or instead of the raw links of the web, from Google to TikTok… but now LLMs might read all the links for us.
ChatGPT and LLMs can do anything (or look like they can), so what can you do with them? How do you know? Do we move to chat bots as a magical general-purpose interface, or do we unbundle them back into single-purpose software?
A conversation with Leonard Brody, Co-Founder and Executive Chairman of Caravan. How do you build brands and consumer products in a world of infinite choice and infinite media? And how does celebrity fit into that? Caravan
Nine months on, everyone is still trying to understand where ChatGPT will go, but one big question for us: how is this useful, for us, today? What's the product? How does this get unbundled?
What is Threads? A Twitter that doesn't suck? Something else? Could it work?
Two weeks after Apple showed us the Vision Pro - what have they built, what is it for, what does it mean for Meta, and why does it cost $3,500? Check back in 2025. Apple's product pages (watch some of this if you haven't already).
Apple does VR. We watched. We took pictures. We talk about it.
We don't know what generative AI will be (or what will happen next week), but we're starting to work out what questions to ask
Buzzfeed News dies just as Reddit and Stack Overflow say they'll charge LLMs to train on their data. Who owns content and how does distribution work in an LLM age? Are those the right questions? What should we be asking?
Crypto crashed, metaverse was silly, and now we know that generative AI is the future of everything. Right? Well, sort of. But though the hype has moved on, the reasons web3 and VR were interesting haven't really changed.
If you spend an hour typing prompts into MidJourney, who owns the result? There are easy answers to this, but they're probably wrong - these are new questions with new puzzles, much like radio, photography or music before.
Generative machine learning is moving so fast it's impossible to keep up. What questions can we ask about GPT4, before everything changes again next week?
The 'ban it' snowball is getting bigger and bigger, but what problem are we solving - privacy, or propaganda? How does this scale to all the other Chinese apps? And meanwhile, how well do we pay attention to the product itself?
Amazon sold close to $40bn of advertising last year - bigger than Prime, bigger than the entire global newspaper industry and probably more profitable than AWS. But is this really advertising, rent, or something else? And what does that mean for Google? Blog post and charts
Microsoft thinks (or says) that Generative ML will reset the search market, unlocking Google's market share and collapsing those 60% operating margins. Really? What would that mean?
What would generative search mean? Generative video? Indeed, Generative products? Last week we talked about how ChatGPT, LLMs and generative ML work - now, what might they mean.
The wave of enthusiasm around ChatGPT and generative AI feels like another Imagenet moment - a step change in what ‘AI’ can do that could generalise far beyond the cool demos. But - it makes things up, and it doesn't actually understand anything it's doing. Probably. What does that mean? What's this for?
Chips have always been the foundation of tech, but the rest of us didn't need to pay much attention - stuff just got faster every year. But now there are actual real, big, interesting structural changes happening - what does that look like?
