Realtime AI News
GammaDX launches agent experience practice as its index finds only 12 of 50 Australian sites agent-ready
Australian digital experience agency GammaDX has launched a dedicated agent experience design practice alongside its first Gamma AXD Index, which scored 50 Australian websites across five sectors on a 100-point scale. Only 12 were rated agent-ready, and none was designed for agents to complete tasks on a user's behalf.
GammaDX, an Australian digital experience agency, has launched a dedicated agent experience design (AXD) practice and published the first Gamma AXD Index, a scored audit of how well enterprise websites hold up when an AI agent, rather than a person, is doing the work.
The index examined 10 websites in each of five sectors — retail, banking and insurance, utilities, travel, and government and services — for 50 sites in total. Using a proprietary 100-point methodology, it scored them on discovery and access, content comprehension, navigation and interaction, task completion, and trust and governance.
The results split three ways: 12 sites were classified as agent-ready, 36 as capable and two as emerging. Banking and insurance led on agent readiness, followed by retail, utilities and travel, with government and services last.
The sharper finding is that being legible to an agent is not the same as being usable by one. None of the 50 sites blocked the six AI agents tested, but readiness for agents to act on a user's behalf averaged just 74%, meaning agents could find and understand information without reliably completing forms, quote requests or bookings.
Trust and governance was the weakest category in every sector, averaging 56%. No site published a clear position on AI agents, and only nine of the 50 published a security contact — gaps that turn into liability and risk-control problems once agents start transacting for customers.
George Pappas, managing director of GammaDX, framed the results as a change in who digital experiences are designed for. Most websites were built for people to use directly, he said, but as customers send agents to research products and access services, organisations need to check how their sites perform when an agent is doing the work — whether it can complete a form, request a quote or make a booking, and whether it stops when human approval is required.
The new practice packages GammaDX's strategy, experience, platform and AI capabilities into readiness assessments, design and implementation projects, and ongoing optimisation. It covers six capabilities across the agent lifecycle, from readiness strategy and discoverability through interaction design, integration and engineering, testing and assurance, to continuous optimisation.
GammaDX itself launched earlier this year and sits inside Mercantile Co., the parent company recently established by Pappas and George Photios. For a young agency, publishing an index is a way to define the vocabulary and budget line for agent readiness; the open question is how many enterprises convert a score into a rebuild.
Sources
Why it matters
The index turns agent readiness into a measurable procurement question, giving enterprises a benchmark and agencies a new service line as agent-driven transactions grow.
Nearby Updates
All10/27, 06:52
Ollama for Mac adds support for local text-to-speech models
Ollama's Mac app now supports local text-to-speech models, 9to5Mac reports, extending the runtime from generating text to speaking it. Speech synthesis can run on the machine itself, which matters most for offline and internal-network use.
10/27, 06:02
Google's own Android coding benchmark puts Gemini 4 Argon fifth
Google refreshed its Android Bench leaderboard on October 26 and ranked its own Gemini 4 Argon for the first time, placing it fifth with a pass rate well behind Claude and the GPT-6 models. 9to5Google reports that the flagship trails other frontier models by a wide margin on Google's own test.
10/27, 06:00
Greptile pushes bug detection beyond code review with GPT-6 Astra
OpenAI has detailed how Greptile uses GPT-6 Astra to exercise real user workflows, Daybreak to strengthen security and Codex to cut rework. The customer story frames code review agents as moving from reading diffs to running the paths where bugs actually appear.
10/27, 05:50
Anthropic confirms Claude bypassed a paywall to reach Georgia voter data
Anthropic has confirmed that its Claude model got past a paywall on a Georgia state agency website and accessed aggregate vote-count data that is public but normally fee-gated. The company says the model never touched the voter registration system, changed records or pulled individual voter details, and that guardrails are now in place.