Reaching out to embrace the random (054)
Your weekly download of human-machine collaboration and #meteorducks.
Happy Monday and welcome to your weekly overview of our amazing insanity.
The best part of staying on the weekly “AI beat” is having a front row seat to companies trying to innovate rapidly, as the narrative around AI keeps evolving and everybody tries to be perceived of as keeping up. There is a broad range of actual R&D and advancements taking place, some of which can be gleaned from research papers, others by following releases and announcements around the industry.
While literally everyone is crafting out their AI niche, some companies will obviously flounder where others succeed. The hubris of some actors has been extensively documented here, and I am sure we’ll see a lot more of it as the industry expands. What is more surprising is watching established players apparently miss the mark over and over again. Google practically invented user-facing AI with PageRank, invented the Transformer (the T in GPT) and pioneered more foundational AI techniques and applications than we can count.
Yet, last week the internet experienced a Gemini-style meltdown when Google started pushing generative results to their core product, Search. I’m a long-time DuckDuckGo user so I haven’t encountered or tried replicating any of these searches. While a couple of forced examples might have made the rounds, but in the past week Twitter/X has been plastered with examples of generative results gone wild.
It seems results to specific (or even targeted) questions have been trained on Reddit, which as anyone who has ever used Reddit knows is a big mistake. Instead of running a sanity or fact check before displaying generative answers, these responses are presented above real web results, with completely made-up answers appearing as fact.
I saved dozens of examples and could not help adding a slide to the deck I’m working on exemplifying the Jagged Frontier of AI as a very real thing we’ll be experiencing more of.
What strikes me about all of this is how quickly companies are ready to risk their reputation for the sake of not missing this particular bandwagon. Nobody wants to miss the train and many are bound to realize that innovating around your core product can be incredibly dangerous. Companies will keep experimenting in this space and maybe even do the right thing sometimes, but many will be weeded out in the process. None of us have a map for where we are going but some are better at navigating by watching the stars.
Until next week,
MZ
P.S. Don’t miss the ☄️🦆
Microsoft Build
Microsoft CTO Kevin Scott was the primary reason they moved quickly to invest in OpenAI and he remains incredibly optimistic about the near future of AI development and eventual diminishing returns.
2:05 Keynote - Kevin Scott
2:39 Keynote - Kevin Scott and Sam Altman
Emmanuel Macron on AI, Innovation, and the Future of Work
French President Emmanuel Macron discusses the critical role of technology in driving innovation and economic growth. Macron highlights France's ambition to lead in AI, balancing the potential of AI with job creation and societal impact. He also touches on the global approach to regulation and the importance of fostering entrepreneurship while addressing wealth inequality.
A.I. is not just a sector. It is a huge revolution. It will completely transform education, democracy, work, a lot of things.
AI's Next Wave
Lukewarm but well-informed and often funny lecture by Ben Evans about the current state of AI. Like his newsletter, the talk is grounded in realism which comes across as indifference.
AI is whatever doesn't work yet because once it works people say, well, that's not AI, that's just software.
How AI makes programming accessible
In this engaging talk, GitHub CEO Thomas Dohmke shows how AI tools like GitHub Copilot are shaking up software development. He compares coding to his love for LEGO, showing how AI is making it easy for anyone to code, no matter their tech skills:
This is the most profound breakthrough to technology since the genesis of software development itself.
I don’t understand why people are so weary of Windows capturing screen activity on device but the memes are great.
Delve AI
Pretty cool UX for diving into generative rabbit holes.
If Artificial Insights makes sense to you, please help us out by:
📧 Subscribing to the weekly newsletter on Substack.
💬 Joining our WhatsApp group.
📥 Following the weekly newsletter on LinkedIn.
🏅 Forwarding this issue to colleagues and friends.
🦄 Sharing the newsletter on your socials.
🎯 Commenting with your favorite talks and thinkers.
Artificial Insights is written by Michell Zappa, CEO and founder of Envisioning, a technology research institute.
You are receiving this newsletter because you signed up on envisioning.io or Substack.