Apple Home AI's Second Camera Adds $240 a Year • The Weighted Average

Apple’s September 9 AI usage-limit document assigns Home features such as video summaries to one camera on the 2TB iCloud+ plan and two on 6TB. At current U.S. subscription prices, moving between those bundles adds $240 a year—a concrete upgrade cost, not a published standalone price for AI summaries.

Unlimited recording is not unlimited intelligence

The important boundary is easy to miss. Apple’s ordinary iCloud+ plan page offers HomeKit Secure Video support for an unlimited number of cameras on the 2TB, 6TB, and 12TB tiers. The new Apple Intelligence support page applies a different limit to some Home AI features: up to one camera, two cameras, and five cameras, respectively. Existing recording entitlement and AI-summary entitlement are not interchangeable. Buying compatible cameras does not by itself buy the right to have all their feeds summarized.

This is useful information for home-automation developers, installers, and households selecting an ecosystem. A customer already paying for enough storage may still face a higher subscription tier to extend a specific AI feature. The decision should therefore start with the number of feeds that need summaries, not the number that can record. Product copy promising broad HomeKit support should not silently imply broad access to every intelligence feature layered above it.

The arithmetic makes the distinction tangible. Apple’s U.S. pricing table lists 2TB at $9.99 monthly, 6TB at $29.99, and 12TB at $59.99. Combine those prices with the separate AI camera-count document: ($29.99 − $9.99) × 12 = $240 in additional annual bundle spending to move from one supported summary camera to two. That is $20 more each month. It includes additional storage and other subscription benefits, so it is not a claim that Apple sells a second AI camera license for $20.

The next step is differently shaped. Moving from 6TB to 12TB adds $30 monthly while the published AI camera allowance rises from two to five. Spread across all three additional supported cameras, the marginal bundle difference is $10 per camera per month. This is an allocation of the bundle price, not a separate tariff. It becomes economically relevant only if the household needs those additional camera entitlements and values the rest of the upgrade appropriately.

Neither calculation determines what Siri AI will cost. MacRumors’ September 9 report says Apple has not disclosed the future fee or how it will be charged. Apple’s own support document says server-side features will have usage limits that can vary with request complexity, demand, policy, and other factors; increased access will be available for a fee. A camera-count table is not a token schedule or a promise of unlimited assistant requests.

There is also a timing caveat. The support page introduces enhanced server-side access in connection with the 2027 software releases, while describing future limits more generally. Its camera paragraph does not establish a complete launch and eligibility schedule. Treat the published table as entitlement guidance to verify before purchase, not as proof that every account, region, device, and software version can use the capability today.

Design the fallback before selling the upgrade

The sensible switch is selective. A household already using 6TB for storage and needing two summarized feeds may have little reason to change its bundle solely because of this document. A 2TB household wanting a second summary feed should decide whether the extra utility justifies the full $240 annual difference. An installer should show both cases rather than presenting the largest tier as an inevitable cost of making a home intelligent.

The Apple Intelligence usage terms are another reason to avoid making an architectural promise out of a subscription label. Developers and integrators should review the applicable terms alongside the feature-specific limits, then document which portions of their proposed workflow depend on Apple delivering a generated result. A summary is a convenience layer; the ordinary way to inspect the underlying recording should remain clear to the user.

For a pilot, choose the feed whose summaries would save the most review effort. Keep a record of missing summaries, incorrect descriptions, and events that still require opening the footage. Those are proposed evaluation criteria, not reported Apple failure rates. The question is whether the feature helps the actual user complete the review task, not whether a generated paragraph looks fluent. Avoid using the presence or absence of a summary as the sole basis for a consequential safety decision.

Commercial uncertainty cuts both ways. Apple’s support page explicitly says supported camera counts and other iCloud+ features may change over time. More generous allowances would weaken the upgrade-cost argument; narrower eligibility would make it more important. The calculation is a snapshot combining two current documents, not a forecast of a permanently fixed price ladder. A future standalone plan could change the comparison completely.

This extends the earlier distinction between iCloud storage and Siri compute access with something more actionable than a possible upsell: a specific feature-count boundary. It also gives consumer-scale form to today’s infrastructure argument about commitments versus usable capacity. Owning the device, paying for storage, and receiving the desired AI service are separate entitlements that need separate checks.

The verdict is to delay a camera-driven subscription upgrade until feature availability is confirmed on the intended account and the pilot demonstrates value. Do not delay ordinary recording merely because the AI layer is uncertain. Evidence that would change the recommendation includes a clearer launch matrix, stable published allowances, or a separate AI option cheaper than the bundle jump. Until then, $240 is the honest full-bundle decision for the second supported summary camera—not an invisible add-on and not a universal price for intelligence.