Microsoft, Amazon, Bard, RunwayML
Just before Thanksgiving
Microsoft is the Real Winner Here
Everywhere I looked yesterday there were opinions on Sam Altman’s departure from OpenAI. I’m not going to point fingers here, but apparently 90% of the 770+ employees at OpenAI threatened to quit and/or demanded the board’s resignation. I’ve previously expressed some doubts about Altman in this newsletter, but I’m going to take this as a sign that he must have been doing something right at OpenAI and this is a deep loss for them and a huge win for Microsoft. Satya Nadella now has a huge investment in whatever OpenAI produces as well as whatever Altman comes up with next. I anticipate a quiet wave of innovation out of Microsoft before spring arrives. The next AI revolution is not flashy toys like Midjourney and RunwayML (more on them later), but boring, practical productivity software to empower the average user. Microsoft is now in the driver’s seat and can leverage this position to gain a huge market share advantage over the other tech kingdoms in their wake.
Why it matters: Microsoft wants to win in the ‘generative AI tools for business productivity’ space and become a dominant name in software again.
Amazon Aims to Provide Free AI Skills Training
I’ve written about free AI certificates from LinkedIn, Google and Microsoft in the past, as well as different paid programs (whose curriculum you should steal). But this new move from Amazon is certainly a wake up call. While it has no relation to the whole Sam Altman drama, it’s a sign of how deep their commitment is have a skilled workforce for whatever their next ambitions are. If you’re still unemployed/under-employed I would seriously consider jumping on this offer. Sure, it’s a bunch of pre-recorded classes on Udacity. Yes, I know that these certificates don’t really mean anything and don’t guarantee you a job. However, I think that they will provide you with the vocabulary and the core principles to be competitive in the job market in the near future. In addition to a bunch of technical courses for developers, they also have entry level classes like these:
Introduction to Generative Artificial Intelligence provides an introduction to generative AI, its applications, and need-to-know concepts, like foundation models.
Generative AI Learning Plan for Decision Makers is a three-course series covering how to plan a generative AI project and build a generative AI–ready organization.
Introduction to Amazon CodeWhisperer teaches participants how to use Amazon’s AI code generator, which produces whole lines of code.
Oh yeah, And they have up to $12 million in scholarships for 50,000 high school and university students from underserved communities globally. Check it out
Why it matters: Amazon wants to train an army of potential employees as it scales up it’s ambitions in the AI arms race.
Bard Sends a Reminder About Homework Help
This is a screen capture from an email I received today. I almost didn’t read it.
While this may have been overlooked amongst all the OpenAI drama, and Bard is well…overlooked, I think it’s pretty important. Google’s tool is quietly and methodically getting better and I think as a Brand, Google still has a lot of credibility. Since I’m a big proponent of AI transforming education, I’m going to be test driving Bard this week and see if it can keep up. My 15 year old son now uses Claude every day and he’s really excited about school and I ChatGPT everyday with my students.
Why it matters: Google is rolling out Project Gemini by the end of the year and wants to become the household name in AI.
The Rundown on RunwayML’s New Motion Brush
Runway released a new feature yesterday which is a big leap forward in AI tools. After uploading an image, you can “in-paint” certain portions and identify those for increased motion. The brush has an adjustable diameter for tight spaces and their is also an eraser tool. There are three sliders for horizontal motion, vertical motion, and proximity. There are also improved controls for zooming, panning, and camera motion. Overall it’s an intuitive interface and solves many of the problems AI tools currently exhibit (I’m looking at you Midjourney). In my initial tests, it was excellent at making the clouds move across the sky and a river flow downstream because these are large, general areas. Unfortunately, it was not any good at making a guitar players hands move or making a car move across a parking lot. Runway still has a lot of distortion and melting that feels like a bad acid trip and weird ramifications that you can’t control. I’m going to do a deeper dive today to see if I can get better results, bit so far I’m not impressed. You can watch a 90 second video if you want to check it out.
That’s it for today. Yesterday, I almost wrote “this will be the only post this week”, but clearly that would have been naive. Maybe I’ll take a break on Thursday and Friday, but since the AI news never stops, neither will I. Since you’ll be mingling. with friends and family over the break and people will ask you, “what do you think about the whole AI thing?”, this would be a great opportunity to spread the word and tell them about this newsletter.