We read the internet so you don't have to.

Hey there!

A lot changed hands in the last few days be it leadership, platforms, or even where customer data actually lives. Individually, each story looks like its own thing. Put together, they're pointing at the same shift.

This edition:

• Three headlines that matter (and why they matter)

• Two myths vs. what we're actually seeing

• One thing worth sharing

Lets go!!

Three headlines that matter (and why)

✦  Apple just changed CEOs for the first time in 15 years.

Tim Cook officially stepped down on September 1, moving to executive chairman, with hardware chief John Ternus taking over as CEO. His first big public moment lands September 9, at Apple's iPhone event where the company is also expected to push its rebuilt Siri AI as a real answer to ChatGPT and Claude.

Why it matters: when one of the world's most valuable companies hands the keys to someone new right as it's trying to prove it belongs in the AI conversation, that's not just a leadership story, it's a signal about how seriously "AI assistant" is now treated as core business.

Source: Apple Newsroom, TechCrunch, September 1, 2026

✦ Salesforce just put its own CRM inside a competitor's chatbot.

On August 26, Salesforce and Anthropic announced "Claudeforce" - a plugin that gives Claude 37 prebuilt sales skills to read live pipeline data, prep account context, and update deals, directly inside Claude, no Salesforce login required. Salesforce's own President of Applications called it a bet that "the UI is the AI."

Why it matters: the company that built its entire business on being the system customer data lives in just admitted the front-end screen might not be where people actually work anymore. If your customer data still assumes someone's clicking through a dashboard to find it, that assumption is aging fast.

Source: Salesforce/Anthropic joint announcement, CNBC, August 26, 2026

✦ OpenAI's ad business just crossed $1 billion, and it's still behind schedule.

OpenAI announced its ChatGPT ads business hit a $1 billion annualized revenue run rate on August 31 up from $100 million in April, in roughly 200 days. It's genuinely fast growth, but it still puts the company well behind the $2.5B it projected for all of 2026. OpenAI responded by opening self-serve ads to India, Europe, the Middle East, and North Africa the same day.

Why it matters: the ad inventory inside ChatGPT is about to get a lot more global, a lot faster than most media plans have accounted for.

Source: OpenAI, Axios, CNBC, August 31, 2026

Two myths vs. what we're actually seeing

Myth #1 : Getting cited by an AI model means the AI is recommending your brand.

Reality: Featured, the company behind HARO, launched a GEO auditing tool this week and shared a specific pattern from analyzing 405 brand audits over the summer - AI engines will sometimes pull a brand's own website as a source, but never actually say the brand's name in the answer it gives the user. So the crawler found you. The user never heard of you. Two completely different outcomes that look identical on a "were we cited" report.

Source: Featured, GlobeNewswire, August 25, 2026

Myth #2: Being cited on Reddit is a durable AI-visibility win.

Reality: A new report from Promptwatch found an 86% drop in ChatGPT responses citing Reddit in mid-August, holding steady for weeks afterward likely tied to a change in how ChatGPT sources its answers. Brands that had leaned on Reddit presence as their main AI-visibility play lost that channel almost overnight, with no warning. Any single-platform AI visibility strategy is one algorithm change away from disappearing.

Source: Promptwatch, via Two Octobers, September 2026

One thing worth sharing

Our co-founder Ankur was on Josh Broward, Ph.D.'s podcast this week, and it turned into a longer, more unfiltered conversation than either of them expected.

Most of it covered ground we don't usually talk about publicly: the gap between the company Ankur set out to build and the one that actually exists, what the fundraise really looked like, mistakes made along the way, and the learning curve of enterprise sales after years in consumer. There's also a good chunk on running a team split across two countries and time zones, and what it actually takes to make that feel like one company instead of two.

Worth a listen if any part of that sounds familiar.