34 – Strong Convictions, Loosely Held

Strong Convictions, Loosely Held: El Prof’s DeFi Predictions for 2025

What’s in a prediction that makes it worthwhile? Is it predicting tomorrow’s closing Bitcoin price? Maybe. Surely the finance people who read this say definitely, but the information arbitrage gap is closing so fast these days that competing algorithms predicting the price create a new chaos unto themselves.

My philosophy when it comes to predicting trends and the future is “strong convictions, loosely held.” I’m easily swayed by compelling new data. Anyone who wants to consider themselves rational has to be — even, especially when it conflicts with our priors, or our ego. (The latter of which is usually the problem when it comes to making accurate predictions in the first place.)

Case in point: Solana is hovering around $180. I said it wouldn’t go below $200. Just goes to show you should not take any advice on financial positioning from me. A qualified investment advisor can coach you on when to buy it. I can just tell you if it’s building towards a real market or not. So, with those obligatory risk disclosures aside, here’s my wildest speculations for what 2025 could have in store.

The obvious potential catalyst is Trump’s looming inauguration. (For more on that, read to the end.) With a leader in charge of the nation’s ego — and as obviously sensitive of his own — you bet your ass leading innovation in the fields of security, cryptography, distributed ledgers and filestore systems, peer2peer protocols, and quantum computing are going to be prioritized over anything else. Why? These are the technologies required to operate as sovereign individuals in an increasingly unified and artificially intelligent world, and that’s good for business.

To that end, here are my official predictions for 2025:

Q1

• Community-managed AI agents with personalities. (s/o @newgleen.eth for inspo on this one.) In other words, personified tokens used to control the weights of different interests that can be held, governed, and stored on a chain. Now that these LLM agent systems have been connected to social media accounts and the like, there are channels for investors to interact with these “characters” and watch them come to life. 

• Crypto payments to creators led by X, followed by Meta, Reddit, and TikTok (if it survives the next couple weeks). The normalization of crypto as social media compensation could in turn stoke the development of utility tokens across all major platforms to reward quality content production programmatically. X already started creator payouts and has been teasing wanting to have a wallet app for years. This one doesn’t seem too far-fetched, or far off.

Q2

• A social credit scoring system. Elon Musk is already flirting with this one on the app formerly known as Twitter, although the logical extreme of leveraging ZK proofs for identity management is more likely to come from Black Mirror rejects like Sam Altman’s Worldcoin project. Let’s grok about that.

• Meta will push their AI agents even further to promote products and engage with its real users. Given the success of the meta narrative of AI agents in general, it’s only a matter of time until the gap is bridged. Maybe it will finally land on a genuinely cool product in the process. I’m thinking one-click AI image generation with the finalized outputs minted to chain as NFTs? Feel free to steal this idea, Zuck. I know you’re no stranger to that.

Q3

• Tokenized brand loyalty programs. Your contact information, the ability to communicate with you, and some basic information about your demographics are worth $10 to $100s per year to brands. That’s how much their advertising calculations tell them their customer retention costs, on a unit economics basis. Imagine capturing a share of that value from the hundreds to thousands of brands for which you’re a potential consumer. The ability to reward your social behavior, which services their marketing and promotion aims, is a more efficient deployment of that capital. With two consistent years of hype behind the web3 space again, loyalty-oriented platforms could emerge to service the major brands interested in the space again.

• First major learn-to-earn platform. On-chain game mechanics are still too slow for the quality gameplay expected of the latest systems. But in pursuit of those game mechanics, retro style games will enable the creation of learning management systems that can leverage the game mechanics to offer a learn-to-earn platform, where a course of information with quizzes can be planned and the completion of levels will be rewarded with tokens. I expect companies to be early adopters, creating novel structures for incentivizing workforce development.

Q4

• Personal finance agents. We’ve already seen some mainstream use cases for agentic management of finances. Bots in web3 are nothing new, but many are a group pot that share and distribute risk. I think the next evolution will be to have a chatbot interface powered by an LLM and your personal data, aimed at helping you build a more aligned financial portfolio over time. (One without the misaligned incentives of traditional financial agents.) With rumors of a tokenized stock exchange in Texas, it’s only a matter of time. What should I buy schwabot?

• Digital ID wallets will emerge as a public utility — like the one we are building here at playhaus. 😏 The data that represents you digitally is valuable to systems like AI. “DID wallets” will allow you to capture that value through chain mechanics that designate you as the originator of the data, not the platform you executed the transaction on, and that will make all the difference. And it’s all built on the back of our buzzword of the week: composability.

—El Prof