2023-01-25 Links

Novel early practical uses for quantum computing, using a technique called annealing. D-Wave is the company that offers a commercial solution.

20 podcasts to listen to over this year, thanks to Ryan Holiday’s compilation.

Hire – and manage – for passsion, says this HBR article.

A fantastic conversation: Carla Harris’ talking with Adam Grant on finding & being a great mentor.

on GPT

A view on the technical difficulty of censoring LLM‘s. I’m not sure I understand this perspective well enough, but if the training data is controlled/limited, wouldn’t the output be limited too?

Google pioneered much of the foundational research that has since led to the recent explosion in large language models.. After the clusterfuck response to its own researchers work, Google is now starting to talk about all the research it has been doing on LLMs.

Tyler Cowen’s latest Bloomberg column  on how quickly LLM’s are getting better.

They are not always reliable, but they are often useful for new ideas and inspirations, not fact-checking.

I’ve started dividing the people I know into three camps: those who are not yet aware of LLMs; those who complain about their current LLMs; and those who have some inkling of the startling future before us.

 

2023-01-24 Links

Toadzilla!

Oh Sh*t! Some use for it, after all?

Lying with visualization: the St Louis Fed Reserve chart on military $ spend

CHARIENTISM [karr-ee-EN-tizz-uhm] – softening bad news with humour; disguising an insult within friendly banter [From Greek rhetoric, kharientismós] Delivering sharp blows with velvet gloves is the epitome of charientism. (HT the wicked David Astle)

on GPT

2023-01-23 Links

Uncomfortable reading for marketing types, from the Economist: How to sell to the young

Just came across Sharon Shannon‘s music today, thanks to John Naughton.

VC Fred Wilson on AI Assist, while he gets back to writing code after years/decades:

The machines replacing humans narrative is powerful. But the narrative I prefer is that AI is making things available that have been expensive and unobtainable for so many. And that is not limited to programming. It is true of so many things.

 

2023-01-21 Links

Wendy Grossman is recording a folk-album 42 years after her first one, & marvels at the way AI is making the process simpler – more impressive, she says, than merely an AI that outputs text in rows (see Nick Cave link yesterday)

Ed Brenegar on generational relationships. Ed’s writing, & the writing of people he’s introduced me to in his network, resonate strongly with me. A simple, yet profoundly valuable question: What should change because of this idea?

Pink Floyd’s playing in the background – Lost for Words caught my attention as I was reading.

To martyr yourself to caution is not gonna help at all

Colin Newlyn’s observations on Jacinda Ardern’s walking away had me nodding furiously (Colin’s writing is to blame for my recent nodding-induced neck-aches 😉 )

GPT:
    • ChatGPT demonstrates a left bias opines  David Rozado
    • Shakoist tries to make the case that humanity’s textual corpus contains enough “models” for LLM’s to progress towards AGI. Fascinating:

Our corpus of text represents the outputs of millions of humans. Embedded within that mess of information is all of our social graphs, the information we observe, and how we have reacted, processed, and written down that information.

 

2023-01-20 Links

HBR;s 8 Trends to Watch in 2023

What makes employees trust AI?

From an interview with the founder of The Museum of Failure, Samuel West:

I realized that there’s no lack of methods or cool processes for innovation. The main obstacles are that people are afraid of failing, and being embarrassed by it—those are two of the biggest obstacles for innovation.

QOTD, James Clear:

“The edge is in the inputs.

The person who consumes from better sources, gets better thoughts. The person who asks better questions, gets better answers. The person who builds better habits, gets better results.

It’s not the outcomes. It’s the inputs.”

GPT:
    • Search through Google Books, and ChatGPT writes a summary from the relevant text: https://www.allsearch.ai/
    • Nick Cave responds to a song written in the style of Nick Cave by ChatGPT

2023-01-19 Links

A long analysis of Elon’s Twitter

Steven Sinofsky on his experience Writing a book on Substack

Common Cog has a provocative, thoughtful piece on why Goodhart’s Law is not as useful as you may think, & some lessons from Amazon’s Weekly Business Review

John Hagel on the Untapped Opportunity of Institutional Narratives.

Benjamin Walker talking to author George P about Listening to Noise.

On GPT:

An analysis of Claude, an LLM from an OpenAI competitor called Anthropic.

2023-01-18 Links

ClaytonChristensen Institute’s primer for understanding the components of a solar energy system for consumers and producers.

On the origins of the phrase “Red Tape

The London Review of Books has this fiery review of a book on consultants by Laleh Khalili.

Raymond Luk on Building a Startup in a Downturn

Nick Morgan asks Why do we deny the facts, and offers some ideas about what we can do about it.

on GPT:

2023-01-17 Links

How to change careers even when you’re super-afraid

Ray Croc: “As long as you’re green you’re growing, as soon as you’re ripe you start to rot.”

Doesn’t this usually happen when a company goes bankrupt? Twitter’s excess furniture sale in progress

 

 

 

Talking points for life: Ideas on how to have conversations (bonus: without long awkward pauses :))

Seth Godin: Generous and selfish

It turns out that happy people are more likely to be generous. (Which implies that generous people are more likely to be happy). Not because they get something measurable in return, but simply because abundance is a choice. And making choices celebrates our agency and potential.

Colin Newlyn offers a few reasons why organisations aren’t shifting to better ways of working?

 

2023-01-16 Links

Everything & more I didn’t know about trucking. I wonder how this translates to other countries?

The differential impact of ChatGPT

any subject area has a core / fundamental area (the science) and an applied area (use of the technique or application repeatedly towards the same or different problems)
– My assertion is that ChatGPT will impact applied area of any topic but not the core area relatively much, where understanding of the underlying building blocks is very important

Bruce Schneir on choosing secure passwords

Surrendering to a drone?