# What Can an AI Agent Change in Google Ads? Full List

Entity by entity: what a connected AI agent can read and change in Google Ads — campaigns, budgets, keywords, ads, assets — and what it never touches.

Questions · Published 2026-08-10, updated 2026-10-26.

> Through a write-capable MCP connector, an AI agent can create search and Performance Max campaigns, ad groups, responsive search ads, keywords, negative keywords, ten asset types and website conversion actions, and can change budgets, bidding strategies, statuses (pause/enable), CPC bids and device bid adjustments, and add geo targets and ad schedule windows. AdCopilot proposes each write first. What it never touches, through AdCopilot by Atromx: deletion of a campaign, ad group, ad, keyword or asset, billing and payment details, and account-level ownership settings.

Vendors love the word "manage" precisely because it is vague. Here is the
unvague version: the entity-by-entity list of what an MCP-connected agent
can read, what it can change, what gates each write, and the column that
is deliberately empty — every row a lever it pulls on your word, in
seconds. The list below is AdCopilot's — 77 exposed Google Ads tools (as of v2.46.25) — and it is the
level of specificity you should demand from any tool.

## The capability table: entity by entity

| Entity | Read | Create / change | Never |
| --- | --- | --- | --- |
| Campaigns (Search, PMax) | Performance, settings, status | Create (paused by default), rename, pause/enable, budget, bidding strategy and target | Delete |
| Budgets | Amounts, pacing, limited-by-budget flags | Raise or lower, campaign by campaign | Delete |
| Ad groups | Performance, structure | Create, rename, default CPC bid, pause/enable | Delete |
| Keywords | Performance, match types, bids | Add, change CPC bids, pause/enable | Delete |
| Negative keywords | Existing negatives at all levels | Add (exact/phrase/broad) to a campaign, account-wide, or to a shared list attached to campaigns — not at ad-group level; take a negative back out of any of the three | Delete a shared list or detach it (done in Google Ads) |
| Ads (RSAs) | Copy, performance, policy status | Create, pause/enable | Delete |
| Assets (ten types) | Coverage, performance | Create sitelinks, callouts, snippets, images and more; link to a campaign, an ad group or the account; pause/enable a link | Delete |
| Conversion actions | Every action, its settings, recent volume | Create a website conversion action (existing ones are never edited) | Delete |
| Google's auto-applied recommendations | Which types are switched on | Pause a type | Switch one back on (done in Google Ads) |
| Geo targeting | Current targets, per-location performance | Add location targets and exclusions | Remove or replace (done in Google Ads) |
| Ad scheduling | Current schedule | Add day/hour windows | Remove or replace (done in Google Ads) |
| Device bids | Per-device performance, current adjustments | Set device bid adjustments | — |
| Billing | — | — | Everything |
| Account settings (ownership, currency, time zone) | Basic account metadata | — | Everything |

Three structural facts sit behind the table.

## Campaigns and budgets: create, adjust, pause, enable

The agent builds a complete campaign — search or
[Performance Max](https://adcopilot.cloud/performance-max-ai), with ad groups or asset groups,
keywords, ads and assets — in one sequence of tool calls AdCopilot proposes first, in the time it takes
to describe it. Campaigns are created
paused by default, so a build cannot spend until a human enables it.
Budget changes and status flips are single-tool writes: proposed with
evidence, live the moment they run, and attributed to your sign-in in
change history; your AI client asks before each one runs if it is set to ask before AdCopilot's write tools. One change deserves more friction than the rest: the agent can switch an
existing campaign's bidding strategy — to Maximize Clicks with a cap on
the cost of a click, Maximize Conversions, a target CPA, Maximize
Conversion Value or a target ROAS. It is proposed like every write, and it
is the one to read twice, since a strategy switch resets learning.

## Keywords, negatives, ads and assets

Keywords add with chosen match types and CPC bids; existing ones can be
re-bid, paused or enabled. Negative keywords — the highest-value,
lowest-risk write in ads — add at campaign level, account-wide or through
a shared list, from search-term evidence. Responsive search ads are created complete (up to fifteen
headlines, four descriptions) and can be paused or enabled; revising copy
runs as create-the-new, pause-the-old — a sequence the agent handles in
one conversation.
Asset coverage — sitelinks, callouts, structured snippets, images and the
rest of the ten types — is often where accounts are thinnest and where an
agent's patience shows first.

## The never column: what is absent on purpose

Three absences define the safety model, and they are absences of
capability, not policy promises:

- **Deletion.** The four remove tools are never exposed, and any mutate
  carrying a REMOVED status is refused server-side in any letter case. In
  Google Ads, removal is permanent — which is exactly why it is
  [not a tool at all](https://adcopilot.cloud/blog/the-case-for-ai-that-cannot-delete) rather
  than a gated one. The one exception is a negative keyword: it carries no
  history and can simply be added again, so taking one back out is a
  write proposed first like the add.
- **Billing.** No tools touch payment methods, cards or invoicing. The
  agent allocates budgets; it cannot reach the instrument that pays them.
- **Account ownership and structure.** Creating or closing accounts,
  changing access, currency or time zone — none of it is in the toolset.

The result is a bounded worst case: everything the agent can do wrong, a
human can put back.

## How a write is proposed, and how your AI client stops it, in practice

The flow is the same for every row of the table. The agent reasons, then
proposes a specific tool call — the campaign update with the new budget,
the negative-keyword add with the exact list. If your [MCP client](https://adcopilot.cloud/google-ads-mcp-server) is set to ask before AdCopilot's write tools, it displays the call and asks before it runs; the change then executes through Google's API, landing in
change history
under your name and in the connector's audit trail with tool, account,
outcome and timestamp. You decline, and nothing happened.

Reads change nothing, so allow them once (in Claude, set the read-only tools to Always allow) and
analysis stays frictionless, which is what makes the
agent worth having. Clients also let you loosen the gate per tool once
trust is earned — negatives on auto-approve, budgets always manual is a
common resting point. The
[approval workflow tutorial](https://adcopilot.cloud/tutorials/approval-workflows-for-ai-ads-changes)
shows the configuration, and
[what an ads agent is](https://adcopilot.cloud/ai-agent-for-google-ads) covers where the
read/write boundary should sit for a new connection.

Print the table, and put the same three questions to any tool you
evaluate: what can it change, what gates each change, and what can it
never do. If you want to see the full list live,
[the free trial](https://app.adcopilot.cloud/signup) exposes every tool above
on the full Pro plan for five days for your first workspace, no card to start (one is needed after 1,000 operations), then, without a card, drops to the Free plan rather than cutting off.

## Questions

### Can an AI agent change my billing details?

No. AdCopilot exposes no billing tools, so payment methods, cards and billing profiles are outside anything a conversation can reach — there is no tool to call. Billing lives in Google's own billing interface under your sign-in. The agent's writable world is campaign management: structure, targeting, budgets, creative — money allocation, never money collection.

### Can it change things without asking me?

Your AI client decides that, and the default answers no: MCP clients such as Claude and ChatGPT surface each write as a permission prompt naming the tool and values before executing, while reads change nothing and can be allowed once (in Claude, set the read-only tools to Always allow). Most clients let you whitelist specific tools for unattended runs — a deliberate loosening you perform per tool, not something the agent can grant itself.

### Does the agent bypass Google's own checks?

No. Every write lands through the official Google Ads API, so Google's validation, policy review and change history apply exactly as they would to a human edit. New ads still enter ad review; invalid values are still rejected; everything is attributed to your sign-in. The agent is a faster hand on the same official levers — not a side door.

---

This is the markdown copy of https://adcopilot.cloud/questions/what-can-an-ai-agent-change-in-google-ads. The index of the site's pages is https://adcopilot.cloud/llms.txt.
