AI2Day Daily Brief — 16 Sept 2026

Good morning, it is Wednesday the 16th of September, and this is your AI Today Daily Brief. The big picture today: a startup selling AI search visibility just hit a one-point-eight billion dollar valuation, Google dropped a mountain of product news, and Washington may leave town without passing AI safety rules that even the AI companies say they want.

Profound, a two-year-old startup, raised 180 million dollars in a Series D round yesterday, landing a one-point-eight billion dollar valuation. Sequoia and Kleiner Perkins led the round, with Lightspeed and Khosla returning. The pitch is straightforward: when a shopper asks an AI assistant for a product recommendation instead of typing into Google, a brand either appears in the answer or it does not. Profound sells software to make sure you appear. Revenue tripled in the six months to September, and the company now has more than a thousand large business customers, including Comcast, Walmart and Estée Lauder. If your job touches marketing or search, this corner of the industry, called generative engine optimisation, is worth watching closely.

Google used a company blog post this week to pull together, in one place, what its AI is doing beyond the chatbot. The headline figure: its products now support more than 300 languages, reaching around 7 billion people, roughly 86 percent of the global population. The health numbers are the ones that stand out. A study run with Imperial College London and Britain's National Health Service found an AI model caught 25 percent of so-called interval breast cancers previously missed across mammograms of 175,000 women. AlphaFold, Google DeepMind's protein-structure tool, has now predicted all 200 million known protein shapes and is used by 4 million researchers in 190 countries. Google's Flood Hub now covers 2 billion people in more than 150 countries. These are not product announcements, they are deployed systems with real use.

Today's sponsor is Zero2Hero: an AI poker coach sits at your table, and you can ask it mid hand whether the player across from you is bluffing. It runs in your browser on any computer. Windows, Mac, Linux, even a Chromebook. Nothing to install. Lifetime access is $99, in US dollars, until 15 September, at zero, the number two, hero, dot A I.

Right, and buried inside that same language push is some engineering that matters more than the headline number. The other 14 percent of the planet, roughly a billion people, mostly live where internet access is absent or unreliable. Google's answer there is a lightweight translation model called TranslateGemma, which runs across 55 languages directly on a phone with no internet connection needed. There is also a voice assistant, built with the non-profit Viamo and powered by Gemini, that has answered more than 2 million questions on basic feature phones in Rwanda. No smartphone, no data plan required. And sign language to text, starting with American Sign Language, is being built into Gboard and Live Transcribe on the Pixel 11. That is language technology functioning as infrastructure rather than a demo.

Also from Google: two new voice models, Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking, rolling out from today. The lighter 3.8 Live model handles 97 languages and can switch between them mid-conversation. The Extended Thinking version is built for multi-step reasoning and scored 82.6 on Artificial Analysis's Speech to Speech Quality Index, taking the top overall spot by that measure. The company says both models can run tools in the background while continuing to talk, offering verbal cues like 'let me check that' rather than going silent. Consumers get access through Search Live and the Gemini app. Paid subscribers also get it in Docs and Gmail. The benchmark is Google's own claim, but the background tool-running is the technically interesting piece: it is what makes a voice assistant feel less like a walkie-talkie.

And finally, Washington. The people running the most powerful AI labs in the world are publicly asking the government to regulate them. That is not a normal situation, and Congress is struggling to respond. The trigger was a researcher resigning from Anthropic and warning publicly that AI could, in that person's words, kill us all by the end of the decade. The chiefs of Anthropic, OpenAI and xAI, Elon Musk's AI company, each made public pledges to slow development shortly after. House Speaker Mike Johnson has signalled Congress will not hold an emergency session before members leave for midterm campaigning. A group of Democrats led by Representative Sam Liccardo wrote to Johnson demanding the House stay. In the Senate, Majority Leader John Thune is working on a bipartisan bill with Senators Klobuchar and Cruz, but it has not been formally introduced. An industry source told CNBC that legislation with real momentum could emerge within the week. Could.

For more depth on all of this, the Monday show AI Today Weekly goes long on the stories that matter most. That is your AI briefing for today. Every story is at A-I-2-Day dot live. That is A, I, the number 2, D-A-Y, dot live. We are back tomorrow morning. Do us a favour before you go: hit that thumbs up, and subscribe for next week.