I published the previous issue of this newsletter on 7 AM Friday and I included some speculation that Apple may spend up to $5 Billion to make up for lost ground in the AI arms race. By 2 PM on Friday Google announced they were investing $2 Billion into Anthropic’s Claude. This comes barely a month after Amazon’s $4 Billion investment in Anthropic and is probably meant as a way to match stay competitive with Microsoft’s $10 Billion investment in OpenAI. If you’d like a better understanding of what this all means, check out this assessment on TechCrunch. Kudos to first year design student Kristopher who referred to it as “21st Century Feudalism”.
The Google investment is just the latest in a developing proxy war between rival companies with a limited number of champions to back. Though all these companies are powerful and expert in many areas of technology, the simple fact is none of them would be able to stand up a credible competitor to either OpenAI or Anthropic in the area of large language models. And since everyone is also betting that LLMs are going to upend their business models and become crucial components of any future tech platform, they can’t afford to not have at least partial ownership of the leaders in the space.
Also in Friday’s newsletter, I discussed the growing alarm over safety and regulation for AI. Apparently, someone has been paying attention, because the G7, the United Nations, and the Biden Administration are all proposing sweeping regulations and safeguards.
Matt Wolfe For The WIN!
While I watch Matt Wolfe every week and he’s been a huge influence on my work, not even I could sit through a 48 minute video comparing all the image generators. If my Medium posts didn’t answer all your questions and you’d really like a detailed analysis of all the major tools, go check it out and support his channel. For the rest of you, I’ll just post this screen cap of the final analysis. Spoiler alert, Leonardo was the winner!
Wei Mao: One Solid Reason to Pay for Medium
So I’m trying to pay attention when I share links to stories on Medium, and avoid those ‘Member Only’ that result in a dead-end paywall for many of you. However, if you decide you’re going to pay for Medium and/or you’ve decided to take the plunge and pay for Dalle-3, you should really check out the work of Wei Mao. He’s written 12 great articles about Dalle-3 in the month of October.
Jakob Nielsen: Jakob’s 2nd Law: Adapt or Lose Job
Yes, that’s right. The OG UX designer and the source of “Jakob’s Law” is back with a very lengthy post about how to get started in AI if you currently work in UX. Fortunately he breaks it down into different sections for Junior and Senior UX designers so you can jump to your section. He offers many excellent suggestions for how to get started and what to be wary of. Here’s the link if you’re ready, or the summary graphics (click to see full graphic).
Geekwire Digs Deeper into Seattle AI Scene
Were you curious about how Seattle Tech and Real Estate companies are using AI in their daily operations? And were you just as frustrated as I was to have missed the Geekwire Summit. Thankfully, they turned one of those talks into a 40 minute podcast and you can listen or read about it here.
“(AI use) …has been off the charts. Really, the amount of innovation and the pace of innovation around AI has picked up so much. The areas where AI has had the most impact are ones that previously involved humans absorbing and synthesizing large amounts of information.” ~Bridget Frey, Redfin CTO
Ready to take the Plunge on an Image Upscaler?
Sometimes I forget that the latest advancements in image upscaler tools is actually driven by AI. If you need large images, AI or “real” (whatever that means) and your workflow requires that you enlarge them to 4K or 8K, you might be a candidate for an image upscaler. Personally, I’m saving up for Topaz Labs so I can use it for video, but if you’d like a full comparison, check out this article from PetaPixel. Spoiler Alert, Midjourney has the best quality, but it only works on MJ generated images.
Dalle-3 is No Midjourney Killer…yet
I’ve been paying for Dalle-3 and ChatGPT+ for a bout 3 weeks now ($20/month) and I’m really impressed with what a great leap forward it is. I use ChatGPT everyday and I’m continuously amazed at how versatile it is. It’s integration with Dalle-3 makes generati g images a breeze and it feels like a quantum leap in usability. As I continue to refine my original image, it just feels like a natural conversation, where I type some variation of “Hey can you do this now?”. While Dalle-3 occasionally fails and I don’t always like the result, their are no weird syntax errors or warning messages like in Midjourney. While I ultimately prefer the images from Midjourney, Dalle-3 has sent a warning shot that usability matters.
This Week’s AI Experimental Film Project
Images in Midjourney, script from ChatGPT, voiceover from Eleven Labs, music from Stable Audio, video from Pika Labs, edited in Capcut. This is just a snapshot in time of what the tools were capable of on this date. Working time, about 4 hours.
That’s it for this week! I certainly hope this newsletter is bringing you some value. Are you getting tired of being the smartest person in the room? Forward it to your friends, so they can be smart like you!