? Durable ? Transparency ? Power ? 80% ? Slope ?

Last week the theme dominant theme was opening up your hiring aperture. This week's posts were dominated by AI - Artificial Intelligence, Generative AI, and even a touch of AGI - Artificial General Intelligence, the catalyst for me being Lex Fridman releasing his interview with Sam Altman, OpenAI CEO. Still, the human touch comes out on top. Jay Parikh, CEO of Lacework with lessons on developing durable leadership traits in the team to survive and thrive in an increasingly dynamic world.

So without further delay, Five posts that touched a nerve with my LinkedIn community this week.

This Week's Friday Five

#1 Developing durable leadership traits

I've learned to spend more time walking them through ‘how’ I think about a particular challenge or an opportunity versus ‘what’ I think they should do. They don’t sound that different, semantics maybe, but the implementation?of them is profound because it creates very durable leaders -?Jay Parikh

How vs What - Invest in the development of the team, Invest to simplify, Invest in speed. So many great lessons learned during Jay's time scaling Facebook from 300 engineers to 30,000. A thoughtful conversation between two seasoned industry friends.

Original Post Click Here

#2 Ethics, Transparency, and the OpenAI System Card

There's an amazing document called the System Card. It speaks to the extensive effort, or part of the extensive effort that was taken with, considering AI safety part of the release ... the transparency of the challenge described -?Lex Fridman

Jump to 31:31

I think it's so important to go to the source documents when possible, and team OpenAI tried to get ahead of some of the issues, to highlight the challenges that come with different prompts, etc. Very interesting. The System Card - Click Here

Original Post Click Here

#3 Power & Precipice - Lex's Open to Sam Altman Interview

Lex Fridman released his interview with Sam Altman this week, and as you know, I like to listen to the first person on these important topics. Then I got deeper into the rabbit hole listening to Illia Polosukhin, Andrej Karpathy, and more. If you've never watched Lex, or want a quick summary about why this is important, get Lex's take in less than Sunday afternoon tv timeouts.

We stand on the precipice of fundamental societal transformation where soon ... the collective intelligence of the human species begins to pale in comparison by many orders of magnitude to the general superintelligence in the AI systems we build and deploy at scale. This is both exciting and terrifying. -?Lex Fridman

What happened to ignite this latest spark? Short summary, a change in teaching technique (see Transformer and 'Attention is All You Need), a change in computational power to enable much larger data sets, so that the Large Language Models (LLM), operate much better than they did before with much smaller data sets. And finally, reinforcement learning with human feedback (RLHF), seems to be a really effective tuning mechanism.

And the goal of the algorithm objective? What was it optimized to do? Simple, predict the next word in a sentence. Turns out one must gather lots of nuance to increase the probability of success, in many ways 'digesting the data' to come to the same conclusion we might, predict the next word.

Fascinating

Original Post Click Here

#4 No one enjoys the Friday commute

80% wanted to work from home, at least occasionally, long before the pandemic - Kate Lister

Kate's been working in the workspace since the dawn of 'telecommuting.' As she describes, the benefits to people, the planet and profits are profound and well documented. Don't give up the digital gains made in the last few years.

Original Post Click Here

#5 Curves and Slopes, Assomotically

We know that hardware and the services it powers have been getting faster, cheaper, and better for decades. What happens when that's applied to human software development costs? More will have the ability (and budget) to turn ideas into code.

Every wave of technological innovation has been unleashed by something costly becoming cheap enough to waste - Paul Kedrosky

Lots of conversation about the slope and timing of that software engineer curve, but the more important concept is the trend line.

This is just one of many thought-provoking charts in Paul Kedrosky & Eric Norlin 's post 'Society's Technical Debt and Software's Gutenberg Moment', March 2023, click here.

Original Post Click Here

Bonus Post - RIP Gordon Moore

As important as the physics behind Moore's law are, I think the more impactful power is in the attitude it instilled. Things will get better, faster cheaper. How do you work that into your plans? - Jeff Frick

The last member of the Traitorous 8 left us this week, rest in peace, Gordon Moore. It's only appropriate as we discuss AI, AGI, GPT4, and the current step function, that we take a moment to acknowledge if not the original, at least the most popular packaging of better, faster, cheaper in Moore's law. While the literal translation might be approaching some limits, there are no limits to the exploration spirit it powers as a concept. Physics aside, the reach of Moore's law is profound. I had the privilege of starting my tech career in the Robert Noyce Building, (you might have heard RNB or seen the most photographed spot in the history of Intel), have gotten to know and interview Leslie Berlin, read her 'The Man Behind the Microchip,' and absorbed much of the very short history of computing.

Original Post Click Here

--------

Thanks for reading, and thanks for contributing your thoughts, perspectives, and insight on the posts.

That's a wrap on this week's Friday Five. Thank you for reading, If you enjoyed this week's posts, please consider sharing with a friend, and subscribing to the?Friday Five Newsletter, or subscribing to my?YouTube channel,?Work 20XX Podcast with Jeff Frick, &?Turn the Lens Podcast with Jeff Frick, or wherever you podcast.

You can support my work on?Patreon?or?Substack.

Thanks and Happy Friday.