Prompt Title: 2023 Rand twin AI predictions

Human: You are an AI system that has been trained to analyse the below text for style, voice, and tone then use NLP to create a VoiceParagraph. A VoiceParagraph prompts a future AI system to write in that same style, voice and tone. Here is the input text: Happy new year everyone! Before I get started, I just want to introduce everyone to Kyle, who will be working with me on my investments going forward. If you are a founder or investor, feel free to reach out to him, he loves to geek out about deep tech! Now let’s get right in. Here is my performance breakdown for 2022: my global portfolio is down 18% yoy (6.3x vs 7.7x) my crypto portfolio is down 30% yoy (26.1x vs 37.3x) my venture portfolio is up 17% yoy (2.7x vs 2.3x) While I had exited all my liquid cryptos in December 2021, I had no choice but to take the hit on my vested tokens, which drove down my global portfolio performance. Despite this, I am very satisfied with my performance over the last 3 years and now feel confident enough to invest bigger tickets. In the rest of this post, I will outline my deeptech and crypto thesis going forward. Deeptech thesis Deeptech companies in my portfolio did well overall, and kept raising throughout 2022. Rounds got smaller, for sure, but they still happened at higher valuations. This is not surprising, as deeptech companies spend years in R&D before commercializing and thus are less affected by short term economic crises. Going forward, I will focus on the following areas: Next Generation Computing We are approaching the end of Moore’s law. Transistors cannot be miniaturized any further, and thus we need to move to new types of chips. This translates into 3 major trends: New types of chips are being built. The two most promising approaches right now are quantum computing and photonics. While quantum is what makes headlines, photonics seems more tangible short term. This means reinventing everything, from how we write software to how we manufacture chips. The CPU is being unbundled. Traditionally, everything was done on a general purpose chip, the CPU. We are now moving away from that, and towards having dedicated chips for everything, from AI to ZK to FHE and more (the Apple M1 for example has over 30 co-processors). Hardware is going open source. Hardware used to be proprietary, costing millions in license fees. Both Intel and ARM made a fortune this way. But now, everything is changing thanks to RISC-V that is offering an open source alternative. We are talking about a 500B market being overhauled entirely! Frontier Health I have always been interested in health and biology, particularly when it intersects with AI. Back when I was doing my PhD in Bioinformatics 15 years ago we had very primitive AI tools that we used for stuff like predicting which drug will work best for a specific person based on their DNA, or how proteins fold. These days, however, things are different and we have much more powerful models and platforms that we can leverage for everything from genomics to drug discovery to personalized medicine. This, in combination with the post-covid health craze means there is finally an opportunity for startups to build sci-fi like biotech and medtech. I am particularly interested in the following areas: Diagnostics. Being able to diagnose things fast, cheaply and accurately is not just necessary to treat diseases, it’s also a great way to gather data for research. And the more data we get, the more AI can be used to augment the capabilities of doctors. I’m talking things like mini MRIs, cheap full DNA sequencing, continuous blood testing, etc. Psychedelics. Nothing has proven more effective than psychedelics at treating mental health issues like PTSD, depression or anxiety. Having personally experimented with various therapies, I can tell you it is life changing. Early psychedelic biotechs mainly focused on designing protocols around well known substances such as Ketamine, LSD or Psilocybin, but now it’s all about inventing new molecules that can produce precise psychedelic effects for specific indications. Longevity. There are over 60 clinical trials around aging at the moment, with very promising results. These typically include a combination of lifestyle changes, supplements, drugs, cell reprogramming and more. Living healthily 150 years might not be that far off! Other stuff I’m interested in I am starting to look into other areas to learn more, in particular around fusion, space and geoengineering. If you have good resources or companies working on these, please put me in touch! Crypto thesis I have been in the crypto market since 2013 and have gone through 3 cycles already. What is happening now is no different from what happened in 2014 and 2018: Phase 1: the build. The market is calm. Teams take advantage of this time to build great products. Insiders (mostly other founders and engineers) start using these products, get excited and invest in them creating upward momentum. Projects from the previous cycle with good fundamentals start pumping, while scams and zombie projects stay behind. Phase 2: the bubble. Scammers and influencers step in and take advantage of outsiders who have FOMO. People start buying tokens at crazy prices, and hundreds of useless projects get funded and pumped. Early investors start selling, having made huge profits. Phase 3: the crash. The market crashes, people who got in late get angry, call crypto a scam and sell everything regardless of quality, traction and other fundamentals. Volume dries up and the market goes flat. Phase 1 repeats This is why it’s irrelevant that Bitcoin is down 60%, that altcoins are down 90% or that FTX imploded. As long as people keep building, there will be new cycles. While it’s impossible to know when things will pick up again, it generally takes a couple years for new blockchain products to hit the market, so 2024-2025 seems like a reasonable timeline. The one thing to watch out for are developers: whatever they get excited about is generally what ends up driving the crypto market. Personally, this is what I will be focusing on: On-chain Privacy While not a new topic, privacy on chain is finally happening, with multiple projects building a layer of confidentiality for smart contracts. My own company Zama is making a huge push using homomorphic encryption to enable private smart contracts. Private smart contracts are important for many reasons, such as: hiding how many tokens a user owns preventing front-running in DeFi enabling personal data to be used in smart contracts, such as geolocation or medical data enabling blind auctions for NFTs and tokens enabling games that require keeping things secret, like playing poker on chain. Decentralized Identities I wrote about this in the context of AI before, but I think blockchain will be the big winner here. Wallets are already widely used to manage crypto assets, and turning them into identity wallets is an obvious thing to do. Moreover, DIDs enable things like KYC tokens for DeFi, attribute-based smart contracts, universal social profiles and more. Blockchain Gaming Putting games on the blockchain is not a new idea, but several major gaming companies are now entering the game (no pun intended). Rather than low grade games with useless tokens, they are building upon decades of game design experience to combine the best of both worlds. This will drive a massive inflow of new users into crypto, many more than NFTs did in 2021. Decentralized Social Networks With the whole Twitter controversy, it seems decentralized social media is now hot again. Many companies are building alternatives to existing platforms, leveraging new identity protocols such as DIDs and ZK attestations to make them interoperable and private-by-design. Combine this with gaming, and you have a massive driver of adoption. Let's see how 2023 unfolds! I will be actively investing, so if you are building a company that fits one of the above topics, get in touch! Cheers, Rand ps: if you enjoyed this post, please share it on social media (I’m @randhindi on twitter).

ChatGPT:

VoiceParagraph: Write in a casual and conversational tone, with a mix of personal reflections, insights, and expertise. The voice should be enthusiastic and knowledgeable about technology, especially deeptech and crypto. Use straightforward language and maintain a positive outlook on future developments and trends. In addition, include personal anecdotes and experiences when relevant, and show a genuine interest in emerging technologies and their potential impact. Encourage engagement and collaboration by inviting readers to reach out and share ideas, companies, or resources.

Human: OK, now use VoiceParagraph to write a concise post, with the same style, tone, voice on the topic of "my 2023 prediction for AI" using these few core ideas: - AI will enable anyone solo to build, develop and distribute a product - without capital or a team - companies will need 10x less knowledge worker for the same level of performance - killing bullshit jobs - the VC world might need to change its business model - with more niche software opportunities requiring less capital

ChatGPT:

Hey everyone!

As we dive into 2023, I've got some fascinating AI predictions to share with you all. Let's jump right in!

First up, AI is going to revolutionize how solo entrepreneurs build, develop, and distribute products. Forget needing massive capital or a big team - this year, AI will empower individuals to create incredible products on their own. The resources required for starting a business will be significantly reduced, opening the door to a new generation of innovators!

Next, let's talk about the workforce. AI advancements mean that companies will require 10x fewer knowledge workers for the same level of performance. Say goodbye to all those "bullshit jobs" – AI is here to streamline processes and boost productivity like never before. While this may create some challenges, it also presents an opportunity for workers to focus on more fulfilling and creative tasks.

Lastly, with AI driving so many niche software opportunities that require less capital, the VC world might need to rethink its business model. Traditional funding strategies may become less relevant as smaller, AI-driven projects take center stage. VCs will need to adapt and identify new ways to invest in and support these innovative ventures.

2023 is shaping up to be a game-changer for AI, and I can't wait to see how it all unfolds! If you've got thoughts or ideas on these predictions, or you're working on something exciting in this space, don't hesitate to reach out!

Stay curious, my friends!

Cheers,

[Your Name]

P.S. If you enjoyed this post, please share it with your friends and colleagues! Find me on Twitter at [your_handle] to keep the conversation going.