During one of our weekly idea sessions, Bob Kernen, jacapps COO, observed that A.I. in this moment in time feels eerily similar to where search engines were at before Google’s algorithm became the default standard for accurate search results at the beginning of the millennium.
Prior to Google’s launch of PageRank, which calculates a website’s relevance based on the other websites that link to it, search engine market share was a battle royale, with many competitors’ engines powered by different technologies.
Yahoo! started off as a manually curated web directory, whereas other sites like AltaVista’s engine automatically crawled the web and indexed websites’ content. As you might remember, search results by platform were so different that many users toggled among multiple search engines to get the results they were looking for. Because of this, new metasearch engines launched like Dogpile, to provide the results of multiple search engines with one click (similar to what travel sites like Kayak.com do today).
Once Google was launched and gained in popularity, the other search engines fell in line: massive consolidation ensued, others shut down, and Yahoo! used Google’s results in their engine. Thus, Google’s dominance has been entrenched ever since.
In the nascency of A.I., what it takes to get the “right” result are pretty identical to the late 1990s in search: using multiple browser windows to run the same query across ChatGPT and Gemini, disappointment, tweaking the query, running again.
But if history is to repeat itself, we should expect this early-stage of A.I. to give way to a frenetic wave of new tools, press releases, and companies; lots of consolidation as engines jockey for market share; and ultimately a new workflow for the world’s online population.
We’ll be here to demystify things along the way. I’d love for this A.I. Edge Newsletter to be the beginning of the conversation. If you have a tip, product, a success (or fail) story, or an idea you’d like to share, please reach out to me: chris@g5j.8ac.myftpupload.com. I’m excited to be on this Ozzy-crazy A.I. train with you.
AI EDGE – TOOLS TO KNOW ABOUT
Audiowriter App: This app takes your phone’s Voice Memos app and adds AI features like voice-to-text and voice clean-up. (Requires ChatCPT paid account). Good for audio recording and transcribing out of the studio.
[It worked well for me and transcribed reasonably well after I got the ChatGPT integration working, which is a little clunky. – CB]
Paste this prompt into the AI engine of your choice to prep for client objections.
*****************************************
Act like a client in the industry listed below who is hesitant about media advertising on the station listed below. Provide 10 common objections to advertising that the client might have and 3 reasons for each objection to convince the client to purchase the ad package.
CLIENT: XXXXXX
CLIENTS CUSTOMER DEMO: GENDER/AGE
MEDIA OUTLET: XXXX
MEDIA MAIN DEMO: GENDER/AGE
CLIENT NEED: XXXXXX
AI EDGE NEWS ABOUT THE MEDIA
NYT Leans Into AI for Ads, Leans Into Humans for News
(Neiman Lab, Axios) On the content side, A.G. Sulzberger states, “In an era in which AI worsens the sort of crisis of trust in digital environments, our advantage is that we are human expert-led enterprises, where journalists are backed by the best editors, and editors are backed by the highest standards.” On the sales side, Axios examines the Times‘ efforts to optimize ads for clients using A.I. while avoiding brand-safety traps.
Universal Music Group Flirts With AI Opportunities
(The New Yorker) Every day 120,000 new tracks are uploaded to Spotify — most of them “functional music” such as white noise. UMG head Sir Lucian Grainge is experimenting with AI and an artist incubator collaboration with YouTube to counter the onslaught.
Reddit Shares Its User Data for $60M
(Engadget) Reddit gets a big check from a “large A.I. company” to analyze its content and user data. Is there a play here for broadcasters with a large spoken-word archive?
AI EDGE ‘WORST PRACTICES’
Air Canada Must Honor Chatbot’s Refund Claims
In 2022, Jake Moffat purchased two next-day tickets on Air Canada to attend his grandmother’s funeral. Before he bought the tickets, the Air Canada A.I. chatbot assured him he could receive a partial refund by filing a bereavement claim — after he purchased the tickets. (Air Canada’s policy actually requires the bereavement fare be approved before the purchase is made.) After being denied the partial refund, Moffat took Air Canada to court, where the judge disagreed with Air Canada’s claim that a company’s A.I. chatbot and a company are “two separate legal entities,” and awarded Moffat $812 and court fees. details
Thanks to Fred Jacobs and Mike Stern for content in this week’s very-human curated newsletter.
Along with his duties at Jacobs Media, Chris develops revenue opportunities for jācapps’ mobile apps clients.
- Ten Recent AI Stories You May Have Missed - November 15, 2024
- The AI Patrol is Ramping Up - October 25, 2024
- All-AI Radio Is Here - October 18, 2024
Leave a Reply