#91 - Search Confusion

Navigating search these days feels like reading tea leaves. Google’s recently launched AI Overviews promise a disruptive future but come with little information and a flawed roll-out. At the World News Media Congress, SEO expert Barry Adams provided some comfort , noting that from the scarce data available, AI changes are not yet drastic. Sites not featured in AI Overviews see a slight drop in traffic, while featured sites could experience a more significant decline, leaving the impression that there is no right way to handle this as a publisher. Google's developer guidelines offer no real help , stating their systems automatically determine which links appear and there is nothing special for publishers to do.

Adding to the concern, during the first week, AI Overviews generated strange and nonsensical results , impacting Google’s credibility. This issue was significant enough that Google’s head of search had to clarify these problems , ironically blaming user queries rather than their own tech. For many in the news media, this feels like a hurried roll-out that disregards their concerns and leaves them guessing about the impact on their traffic and revenue.

Related: Another AI overview analysis seems to confirm the findings of Barry Adams.

Related: Some 2500 secret documents related to Google’s classic search criteria leaked out.

AI

Following other tech companies' announcements, it is now Apple's turn to reveal its AI plans. According to Bloomberg, Apple will introduce a unique approach to artificial intelligence , concentrating on tools that everyday consumers can integrate into their daily routines. The goal is to appeal to users' practical needs, leaving some of the more flashy features to other companies.

Despite the hype, generative AI has yet to become a regular part of people's internet use, according to a new Reuters Institute report on audiences in six countries. For instance, daily use of ChatGPT is rare: 1% in Japan, 2% in France and the UK, and 7% in the USA. However, many believe that journalists are already using generative AI for certain tasks.

Short:

• Financial Times chief executive John Ridding believes news publishers “have leverage and should insist on payment” from AI companies. Link
• Elon Musk’s xAI has raised $6 billion in a single funding round. Link
• Anthropic is debuting chatbots that can answer emails or look for online deals. Link
• Good Wired article on the impact of pocket-sized AI models and how they could unlock a new era of computing. Link

NEWS

This week, the World News Media Congress took place in Copenhagen . Traditionally, this event marks the launch of the Innovation Report , which this year includes a chapter on publishers' AI deals. WAN-IFRA also released a practical report, “AI in Action, ” showcasing 10 global examples of how news publishers are utilizing artificial intelligence, particularly generative AI.

At the event, WAN-IFRA announced the launch of the Newsroom AI Catalyst , a broad-based accelerator program for over 100 news publishers in partnership with OpenAI.

Short:

• Meta walked away from news but is now considering new AI deals with publishers. Link
• OpenAI continues to establish media partnerships: this week Vox and The Atlantic signed a content/data deal. Link
• Not everyone in the industry supports this development. Jessica Lessin, founder of The Information, believes news organizations are absolving AI companies of theft and acting against their own interests. Link
• The Columbia Journalism Review sees no fewer than 10 big questions that publishers should ask themselves when it comes to AI. Link
• A new study shows that generative AI images now make up a significant portion of misinformation images. However, 404 Media claims the paper severely underestimates the problem. Link & Link
• The Journalist Toolbox has started a collection of custom GPT links for journalists. Link
• And Dmitry Shishkin, the new CEO of Ringier Media International shares his vision for a group-wide, AI-assisted user needs strategy. Link

VIDEO

Greg Peters, CEO of Netflix, has an ambitious plan for the company's next competitive advantage: a streaming service that uses AI to adapt even more to customers' preferences. If you like watching documentaries on Thursday nights, you'll be served those, and if you prefer comedies on Sundays, you can expect laughs over brunch.

Meanwhile, the next generation of streamers is getting ready. This week, Showrunner , launched a new streaming service that lets viewers create their own shows using AI.

Short:

• Analyst Evan Shapiro argues the TV space is now dominated by YouTube. Link
• Marion Ranchet bundles her insights from the streaming Media NYC conference. Link
• BBC is starting a review of the representation of “different groups and communities” across its content. Link
• Amazon Fire TV gets a new AI search that lets you chat with Alexa to find stuff to watch. Link
• Only 11% of video ads on social networks are seen in full according to new data from France TV Publicité. Link

SHORT

• Microsoft's bringing the AI of Copilot to Telegram.
• Instagram is training AI on your data. It’s nearly impossible to opt out.
• Reuters Institute Director Rasmus Nielsen will step down in September.
• Sony will use AI to cut film costs.
• TikTok is growing out of Gen Z.
• The Cumulus Media’s 2024 Audioscape report is out.
• YouTube announced the broader launch of Playables, its mobile-style game library.

IDEA

Ever heard of "the egg theory"? It suggests that people prefer AI tools that keep them involved, similar to how adding an egg to instant cake mixes increased sales. Good take by Rex Woodbury on how how user psychology needs to inform AI Products.

a16z believes it's time to reinvent the phone call. A thesis on AI voice agents.

Trauma Aware Journalism is a new news industry toolkit for reporters & newsrooms , with effective tools and best practices to trauma interviewing and preparing for difficult stories.

READ

The 2024 Luma State of Digital report is out. Despite its heavy focus on marketing and ad-tech, it's worth reviewing for its highly visual presentation and concise distillation of key industry trends.

The New Yorker is noticing the rise of a new generation of online culture curators. In a digital landscape overrun by algorithms and A.I., we need human guides to help us decide what’s worth paying attention.

Technological advancements alone aren't enough without considering culture and human desires. Successful design incorporates social and cultural contexts, creating meaningful experiences. Anne-Marie Willis's Ontological Design suggests we shape our world, and it shapes us.

ONE MORE THING

My find of the week. A new Google experiment called Illuminate uses AI to adapt academic content to audio discussions that sound like a radio show.

That's it for this week! Wayfinder is made to travel - feel free to share it with friends & colleagues.

Ezra

?