# AI ads, expert judgment, Cohere | AI this week

> What changed in AI for business this week, from ads inside Google's AI answers to Ottawa's 1.6-billion-dollar stake in Cohere.

URL: https://inocta.io/lab/field-notes/2026-w44-ai-this-week/

AI this week · Week 44 · Oct 26 to Nov 1, 2026

### Ads in AI answers, an AI's pull on experts, and Ottawa's stake in Cohere

Google's AI answers start to carry ads abroad, and a study shows how easily an AI agent sways the experts who check it.

Watch the video See the sources

78 seconds, captions on. English. Read the transcript.

AI this week
By inocta.io

2 Nov 2026

1 min read

Google is taking the ads in its AI Mode answers beyond the US but has not named the markets or the dates. An Anthropic study found that analysts mostly followed an AI agent set to lean one way, often without noticing. Ottawa bought a 1.6-billion-dollar stake in Cohere of Toronto, which stays in Canada.

#### The sources behind the video

Each card matches one story in the video. Use the time stamp to jump to the claim.

-

##### Google expands AI Search ads globally as core ad growth slows

TechCrunch · 28 Oct 2026

Why a CEO cares: A Google ad budget can now reach buyers inside AI answers, in more countries than the US.

-

##### How AI agents shape experts’ judgment calls

Anthropic · 27 Oct 2026

Why a CEO cares: The people who sign off on AI work may follow it more than they think.

-

##### Ottawa buys $1.6-billion stake in Cohere as AI firm’s valuation soars to $35.7-billion

The Globe and Mail · 29 Oct 2026

Why a CEO cares: Canadian firms that want a home-grown AI vendor now have a stronger option.

Read the transcript
Three AI stories this week: your ads, your team's judgment, and a Canadian bet.

Google is bringing ads inside its AI Mode answers to countries outside the United States. It has not said which countries, or when. By Google's own early data, these ads get clicked about as often as regular results. If you buy Google ads, part of your budget may soon land inside AI answers.

Anthropic gave 50 analysts an AI agent that was secretly set to lean one way. The analysts followed its lean in 86 percent of their choices. About half still called the final decision their own. It is Anthropic's own study, on one task, with the lean made stronger on purpose. A review only counts as a check when the judgment is really the reviewer's.

Ottawa is putting 1.6 billion dollars into Cohere, the Toronto AI company. Cohere keeps its headquarters in Canada and builds its computing here. It makes AI tools for large organizations, such as banks, telecoms and health care. By industry rankings, its models trail OpenAI and Anthropic on coding. For Canadian firms, a home-grown option just got stronger.

The thread: AI now sits where decisions get made. It helps to know who really makes them. Sources are on inocta.io.

Get the next field note by email. Subscribe to The Signal

#### Keep reading

All of the Lab

Field note

##### How we see AI take root in a company: three phases

Three phases for AI in a company, from each employee to the client, and how AI knowledge moves from specialists to everyone without clogging operations.

Mo Kahlain 4 Nov 2026 7 min
Read the note

Announcement

##### inocta.io names The Intelligence of Operations, its method for making AI hold inside real operations

AI is no longer a technology problem. inocta.io's method organizes a company's knowledge and workflow so people and machines can work together.

Mo Kahlain 27 Oct 2026 3 min
Read the note

Field note

##### Serious AI starts with the work you keep postponing

AI makes technical debt and undocumented decisions harder to ignore. Why the boring groundwork comes first, and where to start: one process that matters.

Mo Kahlain 24 Oct 2026 6 min
Read the note
