Happy Monday, and welcome to the countdown for an exciting 2025.
Two years ago, we shared the collective epiphany of realizing AI is here.
After playing with ChatGPT and GPT 3.5 a couple of minutes, most of us realized that this is different and that something immensely transformative is about to happen.
As we turn the page on 2024 and reflect on where we are, I suspect many of us feel overwhelmed and barely keeping up with the present – and there is little spare bandwidth to consider how tomorrow might be different. Others see opportunity in the transition and accept the feeling of overwhelm instead of rejecting it. Perhaps by ignoring it, or maybe like treating it as background noise.
How do we maximize the likelihood of collective success in such a transition? What are the means of balancing its negative implications and externalities with its inevitable upsides? How do we not shoot ourselves in the foot with AI, but rather leverage it toward personal and interpersonal growth in every sense of the word?
I suspect quite a few holiday conversations with family will revolve around AI and where it’s going next – so you, their resident expert, will have the opportunity to illuminate them, hear their concerns and work with their fears. If AI is indeed coming for us all, then it is our responsibility to guide its reception.
Until next week,
MZ
Insightful panel from NYT DealBook Summit (90 min)
Do yourself a favor and listen to what this panel of experts thinks is happening in AI and what’s next. It is pointless to try to summarize it, because everyone will have different aha-moments from listening to this frankly awesome session. The lightning round in the end is especially endearing. Bonus points if you spot this issue’s title and its meaning.
Everything you need to know about Sora
Announcement video from OpenAI.
Upskill your coding abilities (52 min)
Hands-on demo and chat with Bolt.new creator Eric Simons by podcaster Greg Isenberg. (Thanks Derek!)
Pre-training as we know it will end
Big claims by Ilya. Apparently pre-training is "done" and what comes next is superintelligence, in surprisingly little detail. Should have been an hour, not 15 min. My impression: as a legitimate researcher, Ilya has little reason to create hype. Maybe I’m being naive.
Revenue breakdown (OpenAI vs. Anthropic)
[
OpenAI and Anthropic Revenue Breakdown
This is a weekly newsletter about the business of the technology industry. To receive Tanay’s Newsletter in your inbox, subscribe here for free…
Read more
a year ago · 41 likes · 3 comments · Tanay Jaipuria
Quick Links
-
I'm fascinated by the point of view of how machine vision sees the world. Here is an app where you upload any photo and have Google Vision API 👀 tell you everything it "sees" in the image.
-
Show & Share: AI Meets Trend Reports - a community showcase of different approaches for digesting a large data set with AI, organized by newsletter friends Iolanda Carvalho and Amy Daroukakis, happening on Jan 15.
-
This open source library by Microsoft is incredibly useful if you want to extract text from pdfs, docs, images, audio, etc.
-
Vanilla Backrooms: I am staring to wrap my head around the infinite backrooms experiment by @andyayrey. Thanks Chris!
-
I spent 8 hours testing o1 Pro ($200) vs Claude Sonnet 3.5 ($20) - Here's what nobody tells you about the real-world performance difference. Interesting test cases from Reddit.
-
This is really cool, an AI tool to navigate art by vibes (or the meaning and semantics of text) featuring art from the Amsterdam Rijksmuseum. Read more.
-
Lots of QOL upgrades for those using Google NotebookLM. Most interestingly they're now doing interactive audio, so you can apparently interact with the podcast hosts. Read more.
-
How to build a useful AI product, by two of my favorite makers - Granola CEO and Every.
-
Gorgeous essay illustrating how LLMs can be used to expand our perception.
Dystopian agentic vibes (via X).
You are receiving this newsletter because you signed up on envisioning.io or Substack.