Exploration #118

Hate Longscrolling? Jump Around!

• Focus…

• Learn…

• Think…

• Know…

• And finally…

Hi all. This week we’re looking at SearchGPT and the explosion in generative search, Microsoft makes security everyone’s job, and a +1 from me on Evan Shapiro’s interview in Current.

But First…

We’ve got quite a few announcements and opportunities this week. Without further ado…

AI at the NETA Annual Conference 

The breakout sessions for the NETA Annual Conference in Pittsburgh were announced last week, and I’m happy to report that we’ll be presenting our session, “Everything That’s Wrong with AI” as part of a group of sessions curated by Executive Content Managers’ group. Amber, David and I appreciate their support. And don’t let the cheeky title fool you into trying to get torches and pitchforks past TSA en route to Pittsburgh. Our intent is to lean into the trough of disillusionment into which generative AI is currently dropping, but also to respond to feedback we received from our PBS session on AI and talk about the ethical considerations that come with using these tools.

Amber and I are also going to be participating in the session “Getting the AI Conversation Started at Your Station“ produced by Suzanne Smith at WFSU and Jessica Solberg at Idaho PBS. You can send a copy of your draft or completed policy to Suzanne at [email protected]

AAA Webinar Opportunity (Adobe, AI, August) 

Additionally, August 15, at 1pET/10PT will be our next 3rd Thursday webinar: Adobe + AI: Practical Applications for Public Media, featuring Brooke Hopper, Principal Designer of Machine Intelligence & New Technology at Adobe. 

In this session, Brooke will delve into the current state of generative AI in the creative fields, exploring the transformative impact of machine learning on the media industry. Attendees will be treated to a detailed demonstration of various use cases within the Adobe Creative Cloud ecosystem, showcasing the practical applications and innovative potential of these cutting-edge technologies. 

Additionally, Brooke will provide valuable insights into how to ethically and effectively integrate these new tools into your workflow, addressing common challenges and offering best practices. This webinar is not a tutorial in how to use Adobe AI tools but is an opportunity for you to stay ahead of the curve by learning from one of the leaders within Adobe. 

We first encountered Brooke at South by Southwest during her session, Creativity in Flux: Art & Artificial Intelligence. You can listen to that session here, and interact with Brooke directly next month. Register now!

March of Progress

Speaking of South by Southwest and registrations, registrations opened Tuesday, August 6 for South by Southwest Interactive, March 7-15, 2025. In addition to the standard early bird discounts, you’ll also get a better choice of hotels if you register as close to the opening of registration as possible. This means being able to choose a hotel that finds the sweet spot between room rate and transport to/from the conference activities. Around 30k people attend this conference, so there are definite advantages to getting your name in early. This will be my 12th ‘South-by,’ so you are welcome to email me at [email protected] if you’d like guidance on this or on making the case to your leadership for attending next year.

We had a number of stations represented there last year as well as folks from PBS. And we’d love to increase the size of the community in attendance this year. I can tell you, unequivocally, that the contacts I make and the things I learn at ‘South-by’ pay dividends for the rest of the year between conferences. Learn more here.

Evan Shapiro is Right 

A decade or so ago, I head someone say, “What got you here, may not get you where you’re going.” I’ve been thinking about that phrase a lot in the last couple of years as I think about the future of public media, and also often when I hear Shapiro speak. And that’s why I’m making his interview with Current the first piece I’m highlighting in what is now called our “Focus” section (fka “If You Click Only One”). Scroll on….

Focus…

Why public media needs to let younger decision-makers call the shots (Julian Wyllie - Current) - If you've already read Wyllie's interview with Evan Shapiro, it's worth another look, especially if it made you uncomfortable the first time. I agree with Shapiro's assessment that only the NYT and BBC (and maybe Channel 4 in the UK) are the only legacy brands that are truly embracing the changes that will be needed to survive in the future. We cite them here frequently. I also agree with Shapiro's assertion that we're not ready and that the right people are not in the room to get us ready. I'm fortunate enough to have a young team in Labs. I think the average age is under 25 and the oldest of us are Gen-X. I learn a lot from them. Their values very much align with public media's values, and they seem to appreciate being able to spend part of their careers working for public media. But by my estimation, 90% of what we create as public media isn't relevant to them. If you believe public media should exist after you retire (whatever your age), that should scare you.
—And lest the radio folks feel left out, Fred Jacobs released “Is Public Radio a Victim of Its Own Org Chart” (Part 1 and Part 2) last week. His point-of-view, which swings low with that old time content religion, made me a bit nostalgic for my TV programming days: “I believe a reality of radio (or television or film or podcasting) is that the central direction of a media organization’s content model has to emanate from the vision of one person.”

Learn…

AI is confusing — here’s your cheat sheet (Jay Peters - The Verge) - It's been a while since I've seen a good primer on AI. Even if you've see some of the others I've posted in the past, this one is a great review. Reading it, I realized I had started conflating "foundation" models and "frontier" models in my own daily chatter. So, there's definitely something in this one for everyone.

Think…

Why The Atlantic signed a deal with OpenAI (Nilay Patel - The Verge) - The editor-in-chief of the verge interviews the CEO of The Atlantic on why and how a 167-year-old publication is embracing generative AI. There's something here to consider for anyone who leads or wants to lead a public media company in 2024.

Is AI About to Run Out of Data? The History of Oil Says No (Niall Ferguson & John-Clark Levin - Time) - Here's an interesting perspective that digs deeper into the "data is the new oil" metaphor. There are pieces in the next section about AI companies paying for access to content, and otherwise scraping websites (sometimes to excess). The reason they are, in some instances, unscrupulously scraping sites is because they need cheap data for training purposes. Read this piece first, and those will take on a new light for you.

Copyright Office tells Congress: ‘Urgent need’ to outlaw AI-powered impersonation (Devin Coldewey - TechCrunch) - Should you have similar protections again unauthorized digital replication as the content you or your teams make at the station. The US Copyright Office makes a compelling case in the first of several reports about copyright and AI.
—Or, jump straight to the US Copyright Office's Copyright and Artificial Intelligence page.

Dear Google, who wants an AI-written fan letter? (Anthony Ha - TechCrunch) - As with so many things Google has put out into the culture (looking at you Google Glass) this one is ahead of its time. Eventually, I do think today's young kids, who are growing up with AI tools, will work with those tools to help write all sorts of important communications. If anything, doing so won't be worthy of an ad, because it will just be part of daily life, like using a calculator instead of doing long division by hand. So maybe Google failed to read the room, but I see the backlash as a reaction to that tone deafness, and not something that will ultimately stop the use of this technology.
—Eventually, as Andy Stout reports in RedShark News, the ad was pulled.

Know…

Every Microsoft employee is now being judged on their security work (Tom Warren - The Verge) - While this is more the purview of our sister peer learning community (PLC), Cyber, in the spirit of Microsoft’s move, I’m sharing it with you as well.

OpenAI announces SearchGPT, its AI-powered search engine (Kylie Robison - The Verge) - When ChatGPT launched at the end of 2022, it was common to use the tool as a search tool, often to regrettable or even embarrassing results. Of course, that’s not how the tool was intended to be used. But that all about to change. As Google and Perplexity continued to push into “generative search” OpenAI was developing their own tool.
—And, not that I'm surprised, but the OpenAI demo was not flawless, as reported by Matteo Wong in The Atlantic: OopsGPT
—See OpenAI’s own announcement for their SearchGPT Prototype 
—Or jump straight to the SearchGPT waitlist.

Bing previews its answer to Google’s AI Overviews (Kyle Wiggers - TechCrunch) - Not to be outdone by SearchGPT, Microsoft has officially launched generative search via Bing. Of course, that was the promise a year ago, and after the initial blush everyone realized that Bing was still, at its core, Bing, and went back to Googling things.
—Microsoft's Announcement: Introducing Bing generative search

Microsoft confirms Reddit blocked Bing Search (Barry Schwartz - Search Engine Land) - As nifty as Bing's generative search might be, "content is king" and if you're being blocked from key internet content, AI magic is only going to take you so far. But one does wonder how this will play in the wake of this week's antitrust news. Of course, Reddit licensed Google the rights to exclusively crawl the site, but then dropping stacks of cash on companies is what earned Google the US Justice Department spotlight in the first place.
—Of course it's not just Bing that's impacted, as Emanuel Maiberg reports in 404 Media: Google Is the Only Search Engine That Works on Reddit Now Thanks to AI Deal

Perplexity announces revenue sharing deal with publishers (Reed Albergotti - Semafor) - Search engine upstart, Perplexity, is working to settle its legal issues with some good ol’ fashioned revenue sharing. I still think we should be at the table to benefit from the content we create or help create.
—And, as Sara Fischer reports, Media giants strike new AI deals even with minor up-and-coming generative search players.

AI's search quake shakes media landscape (Sara Fischer + Scott Rosenberg - Axios) - Axios gives us a good summary of where many of the AI players stand relative to search. Google isn't mentioned, so we have to just assume we know it's their hill to defend. Of course, that'll be more difficult now that, as Nadine Yousif reports for the BBC, they've been found guilty of illegally maintaining a monopoly.

Meta releases the biggest and best open-source AI model yet (Alex Heath - The Verge) - There was a lot of buzz about this model release from Meta in the past couple of weeks. Of course it’s not a packaged product, like ChatGPT, Midjourney, or Suno. So, you’d need to be a developer to get anything out of it. But it is a very deliberate shot across the bow of other AI companies. And, unlike the proprietary products from those companies, Meta is making it available for all. Competition is good, but, as I always say, beware Geeks bearing gifts.
—See Meta’s announcement at: Introducing Llama 3.1: Our most capable models to date
—And read the party line on open source tech at Meta, in Mark Zuckerberg’s essay, Open Source AI Is the Path Forward
—Also of potential interest is this hour-long live, fireside chat between Jensen Huang, CEO of Nvidia, and Mark Zuckerberg, where they touch a bit on Zuckerberg’s open AI philosophy.

Adobe rolls out more generative AI features to Illustrator and Photoshop (Jess Weatherbed - The Verge) - Some update from Adobe that could be topics for discussion at our August 15 webinar with Adobe's Principal Designer of Machine Intelligence & New Technology.

The Backlash Against AI Scraping Is Real and Measurable (Jason Koebler - 404 Media) - While Nebraska still does not block crawling by AI bot (with one exception), I know many of you do. And you are not alone.
—Whatever your stance, as Koebler also reports: Websites are Blocking the Wrong AI Scrapers (Because AI Companies Keep Making New Ones). So, if you are choosing to block scraper bots, make sure you are blocking the right ones.

Anthropic’s crawler is ignoring websites’ anti-AI scraping policies (Jess Weatherbed - The Verge) - Not that blocking scraper bots does a lot of good if the bots ignore the block. I've cited this here before. Claude, the world's nicest chatbot, has a dark secret.

Using the term ‘artificial intelligence’ in product descriptions reduces purchase intentions (Eric Hollenbeck - WSU Insider) - I continue to be amused by how quickly we've gone from a time when AI was a marker of something magical or superhuman to a time where AI is a shorthand for a subpar product (be it an app, music, or a piece of writing).

GOP’s ‘Project 2025’ Eyes End Of Public Media Funding, Non-commercial Status. (Inside Radio) - While Project 2025 has been marginalized since this piece was written. It's useful to know the specifics of the authors intentions. When things like these come back, they tend to come back quick.
—Roxana Muenster also has a good breakdown of the threat to public media in Brookings Institute’s Project 2025: What a second Trump term could mean for media and technology policies. Look specifically for the section: “The Media As the ‘Enemy of the American People.’”

And finally…

Autobiographer - And finally, here's an interesting new app for those thinking about legacy. I'm not endorsing it, but its existence is a sign of the times. I do think this could be hugely influential for some people (and their families). But for me, who's been intending to write a diary/journal since I was 10, I know I wouldn't stick with it.

Have a creative, productive week!