AI Visibility Checks Now Run on Current-Generation Engines

In short: WriteWorks has refreshed the AI engines your prompts are measured against, so every visibility check now runs on the current generation of each model - including GPT-5, Gemini 2.5, Grok 4.1 and Llama 4. Your mentions, citations and sentiment scores reflect the models real people are using today, not versions that have since been retired. It's aimed at growth and search-marketing teams who report on AI-search visibility and need those numbers to hold up under scrutiny.

Model line-ups shift quickly, and a monitoring tool pointed at last year's versions will keep returning results without ever telling you they're out of date. This update removes that silent drift.

What is the current-generation AI engine line-up?

The current-generation AI engine line-up is the refreshed set of models WriteWorks uses to run your AI visibility checks, with each engine pointed at its newest available version rather than a legacy release.

When you run a scan, WriteWorks sends your prompts to each engine and records whether your brand is mentioned, which sources are cited, and how the response frames you. Which model answers that prompt matters - different generations retrieve different sources, cite differently and describe brands differently. Measuring against the current generation means the results match what your audience actually sees.

How the refreshed engine line-up works

Every engine in WriteWorks routes through a single AI gateway, which is what makes a line-up refresh like this possible without any change on your side.

The work behind this update was deliberately unglamorous:

• Each model was verified against the live catalogue to confirm it is still available.
• Versions that had been discontinued were retired from the line-up.
• Every engine was pointed at its newest equivalent - GPT-5, Gemini 2.5, Grok 4.1, Llama 4 and others.

Because the gateway sits between WriteWorks and the engines, your saved prompts, scan schedules and historical reports stay where they are. There is no re-onboarding, no re-authorisation and no configuration step to complete.

A useful side effect: several of the current-generation models cost less to run than the versions they replace, so scan volume goes further.

Who this update is for

This update matters to anyone who treats AI-search visibility as a reportable channel rather than a curiosity. It's especially useful if you:

• Report AI visibility, share of voice or sentiment to a leadership team and need the underlying data to be defensible.
• Compare your brand's presence against competitors across several engines and want that comparison to be like-for-like on current models.
• Track citations to decide which pages and sources to strengthen, and need those citations to reflect current retrieval behaviour.
• Run scans across multiple engines and care about the cost-per-scan of doing so at volume.

Advanced users benefit too. If you maintain large prompt sets across markets or product lines, a stale model in the line-up quietly contaminates every trend line built on it - and the more prompts you run, the harder that is to spot by eye.

What this changes for your reporting

The practical shift is that your visibility numbers now describe the present rather than an approximation of it. Nothing changes in how you work; what changes is how much weight the output can carry.

Three things follow from that:

1. Trend lines need a note. If you are comparing results before and after this refresh, expect some movement that reflects the model change rather than a change in your own visibility. Annotate the switchover date in your reporting so the shift is explained rather than investigated.

2. Engine-level differences become more meaningful. When every engine is on its current generation, a gap between two engines tells you something real about retrieval and sourcing - not just that one was running an older model.

3. Recommendations age better. Content and PR decisions based on which sources get cited are only as good as the model doing the citing.

How to run a scan on current-generation engines

There is no setup required. The refreshed models are already in place for new scans.

1. Open Analytics → AI Visibility.

2. Run or schedule a scan exactly as you normally would.

3. Review your results - they now come from the latest model for each engine.

If you keep scheduled scans running, the next scheduled run will use the current-generation line-up automatically.

Tips for a better result

• Re-run a small, representative prompt set first so you can see how results move before your next full reporting cycle.
• Mark the switchover date in any dashboard or deck that shows a long-running trend, so month-on-month changes are read correctly.
• Review prompts written for older model behaviour - phrasing that was tuned to work around a previous generation's quirks may no longer be necessary.
• Check your engine selection. If you narrowed the line-up previously because a model felt unreliable, it's worth reassessing now.

Frequently asked questions

Which AI engines does WriteWorks now check against?

WriteWorks measures prompts against the current generation of each supported engine, including GPT-5, Gemini 2.5, Grok 4.1 and Llama 4, alongside the other engines in the line-up. Each engine was verified against the live model catalogue, with discontinued versions retired and replaced by their newest equivalent.

Do I need to change anything to use the current-generation engines?

No. The refresh applies automatically to new scans, including scheduled ones. Your saved prompts, engine selections and historical results are unaffected. Open Analytics → AI Visibility and run or schedule a scan as usual - the results will come from the latest model for each engine with no setup step.

Will my visibility scores change after the refresh?

They may. Different model generations retrieve and cite different sources, so mentions, citations and sentiment can shift even when nothing about your brand has changed. Treat the switchover as a known event in your trend data and annotate it in reporting, rather than reading early movement as a performance change.

Does this affect the cost of running scans?

Several current-generation models cost less to run than the versions they replace, so scan volume goes further than before. Exact usage and allowances depend on your plan - see WriteWorks pricing for details.

Which plans include the current-generation engine line-up?

(Confirm before publishing - plan availability to be verified by the WriteWorks team.)

The bottom line

Measuring AI-search visibility against retired models produces numbers that look fine and quietly mislead; running every check on the current generation of each engine keeps your mentions, citations and sentiment tied to what your audience actually sees.

If you'd like to see how your brand shows up across current-generation models, open AI Visibility in Analytics and run a scan - or browse the WriteWorks resources hub for more on tracking brand presence across AI engines.