Inklings #010 📧
While I understand the attempt at an AI-first computer, there are a few strange things about the 'Googlebook' starting with the name on down. Of course, that could just be because much of it is TBD at the moment, perhaps related to Google IO kicking off next week. Still...
M.G. Siegler
Thoughts On...
🛍️ Amazon.com Adds Alexa – I never fully understood why Amazon created a separate brand/product for augmenting their core shopping experience with AI. I guess (and guessed) it was to be able to experiment with 'Rufus' while at the same time, the earlier versions of Alexa may not have been up to ruff snuff – especially if you're going to be putting it in front of millions of people instantly. With Alexa+ now ready and rolling (out), it's seemingly time see if Amazon's AI can be used to buy the big boy pants. And presuming it works (Rufus has gotten much better over time in my own usage), this could help Alexa itself, giving Amazon an advantage in the all-important AI commerce – and eventually agentic commerce, space. [Verge]
🎥 AI as a Filmmaking Tool – You can put Peter Jackson in the James Cameron camp with AI. That is, rather than freaking out about it, simply viewing it as yet another tool to leverage (or not) when it comes to making movies. It is interesting how the camps at least somewhat seem to be dividing between filmmakers who favor special effects versus those who favor practical effects... I'm not saying AI is simply the new CGI, but well, Jackson is with: "It's no different from any other special effect." He goes on to note that he's not talking about AI in general, and that he's obviously in favor of IP and identity rights. But again, it's just the common sense: if you can leverage something to better fulfill your vision as a filmmaker, why wouldn't you do that? Because people aren't as involved as they are in say, stop-motion filmmaking? What is the cut off line we're drawing here, and why? To save jobs? Jobs have been going away in Hollywood for years, long before AI came around. There will be new jobs to leverage the new tools, just as is always the case. Also, if and when fully AI-generated movies come, I suspect once the novelty wears off, very few people will want to watch them. And the human-created ones (perhaps augmented by some AI tools) will rise in value. [THR]
📉 Is Microsoft Now Undervalued? – The company has just sunk below $3T in market cap. They're $2.6T behind NVIDIA and Amazon is about to overtake them to push them into the 5th most valuable company in the world spot. About a year ago, I made the case that Google was severely undervalued relative to their parts. The company was at $2T back then – ahem – now they're at $4.8T. Microsoft's story is a bit different since their parts are a bit more downtrodden at the moment (Xbox, LinkedIn, etc). Still, it's hard to argue with their continued results, quite literally. And with that 25%+ stake in OpenAI. If they can actually figure out AI on their own, obviously the stock will bounce back. But Wall Street is clearly also skeptical of that at the moment and the massive CapEx spend. (They're also obviously skeptical of Apple and AI, but with next-to-no-spend, they're near an all-time high at $4.4T.) Are activists going to start circling like they did at the end of the Steve Ballmer era? No one is going to force Nadella out, but might they force some spin-outs? [Information 🔒]
I Wrote...
M.G. Siegler
I Quote...
"OpenAI is Coke, Anthropic is Pepsi and Grok is RC Cola. I never really saw people drinking it."
– Ben Pouladian, an engineer and investor, on the notion that growth at Grok, xAI's chatbot, has stalled. While the racy stuff may have worked for a bit, I wonder what percentage of usage is directly tied to X, and even more specifically people asking it to fact-check tweets (something which Meta is on the verge of copying with Threads, as is their way).
While forcing banks currying favor ahead of an IPO to use it is certainly one strategy, that's not really scalable either (nor does that appear to really be working). So what gets Grok back into the race? Anything? Simply throwing money seemed to work at first, then didn't. Maybe tying it directly to Cursor if/when they complete that full deal?
Below, members of The Inner Ring will find thoughts on:
• Who Is Winning Elon v. OpenAI?
• OpenAI's $7B Man With His Own AI Lab