Andrew Bruce Smith

💬 IBP Summer Party and AI Panel Debate - 17 July

AI in journalism & comms: how far is too far? A lively discussion chaired by @andismit.bsky.social‬. Panel: @paulbradshaw.bsky.social‬ @sophiasgaler.bsky.social‬ @mrkennycampbell.bsky.social‬ @bysianharrison.bsky.social‬
Register:

Bizarre thing happened to me yesterday that my teenage self would not believe. I was playing at the gig to celebrate the life of the genius that was Neil Innes. Whilst waiting backstage to go on, Michael Palin sang the lumberjack song to me, in German.

#OnThisDay 3 years ago, @cipr-global.bsky.social’s The AI & Big Data Readiness Report was lauched from our #AIinPR panel. Great research that demonstrated the huge potential of #AI & Big Data for #PR early on-lead by Prof Anne Gregory, Swati Virmani & @andismit.bsky.social cipr.co.uk/CIPR/CIPR/Le...

Interesting. But can we really replicate the behaviors of humans just by using AI agents plus their answers?

I think that's pushing the line of what can be simulated too far, while losing the central point that LLM are just very complex predictors the next word.

The Reuters Institute is seeking a new lead author for the Digital News Report Next year’s edition will be the last one led by Nic Newman, who launched the report in 2012. Here’s what the role requires and how to apply.

🚨 Big news today at the Institute

We are seeking a new lead author for the Digital News Report

Next year’s edition will be the last one led by Nic Newman, who launched the report in 2012. Here’s what the role requires and how to apply
reutersinstitute.politics.ox.ac.uk/news/reuters...

The most comprehensive online community glossary in the world — Michelle Goodall The biggest and best free online community glossary. Let me help you demystify the terminology used in community strategy.

Back to #Community stuff y'all.
Here's a useful A-Z of Community glossary (free) I created www.michellegoodall.co.uk/insights/onl...
[fun fact: my 1980s vinyl record cases have this pattern]

The growth in this place feels extraordinary. Could be one of the big media stories of the year.

Elon can tweak the algo to promote the dickheads as much as he likes, but he can’t stop (and arguably, will accelerate) people leaving in droves because it’s just not a fun place to hang out anymore.

Network effects work both ways

Just listened to a 10 minute Notebook LLM podcast on my new series The Moderators based on the article I wrote for the BBC website.
It was conversational, if a bit slow, and it included everything I wrote plus a couple of extra bits.
Don't think I'm out of a job quite yet - but it's impressive

🗞️ More and more Times and Sunday Times journalists and staff are joining BlueSky.

We’ll be updating this starter kit as they do so, you should give them a follow

👇

Navigating the Nuances of AI with Anthropic’s Philosopher in Residence: Amanda Askell As someone who has studied and maintained an interest in both philosophy and AI for over 40 years, I admit to being a bit jealous of Amanda Askell, Anthropic’s head "prompt whisperer", who also has a ...

The Lex Fridman interview (@lexfridman.bsky.social) with Amanda Askell at Anthropic yesterday was very interesting on many levels - I had a few thoughts on the back of it - but you’ll have to go off to LinkedIn to read it. Let me know what you think.

www.linkedin.com/pulse/naviga...

Study: Growth of AI adoption slows down among U.S. workers A survey finds the growth in AI use in the U.S. workforce was nearly flat over the last three months.

Also this...

www.axios.com/2024/11/12/a...

“Slack also found that globally, nearly half of workers (48%) said they were uncomfortable telling their managers they use AI at work.

Among the top reasons cited were a fear of being seen as lazy, cheating or incompetent”

Yes. ChatGPT, etc in their retail versions are all personal productivity tools. They aren't designed to be used in a collaborative fashion. A good prompt shared could benefit other colleagues, for example. But in corp cultures that don’t incentivise that behaviour, people will hoard their assets.

Also people are competitive. Data presented at a conference I was at recently showed a significant % of ppl using AI to be *personally* more effective - to get stuff done faster or better than colleagues - and not admitting as much publicly.

Morning, morning.

Thoughts after 24 hours: Twitter will be just fine.

We’ll instead have an ideological imbalance both here and there, with some (myself included) bouncing between, for now. Is Bluesky the ‘left’s’ Truth or can it become peak-era (pre-driven by ideology either way) Twitter?

Worth watching this social sorting that’s going on. The Twitter X-Odus isn’t the only thing spurring Bluesky’s growth by 700k users a week. Threads’ algorithm and user matching is pushing its users to try Bluesky too. Is Threads still growing by 1m a day? I bet it’s now more.

This place is buzzing! Hello if you’ve just joined and found me - I’m the BBC’s technology editor 👋

And new at number one.....

I’m cautiously hopeful that this might turn out be a decent venue 😀

a video of a man waving with the words greetings written below him ALT: a video of a man waving with the words greetings written below him

So this is where everyone is 😀