Nvidia and Foxconn Push Agentic AI Into Taiwan Hospitals

Article summary

Nvidia and Foxconn are working with Taiwanese medical centers on agentic AI systems for clinical and hospital operations. The effort is tied to Healthy Taiwan and a USD 1.5 billion sovereign AI healthcare investment. CoDoctor, CoDoClaw, Scrub Bot and Nurabot show healthcare AI moving toward multi-agent and physical AI workflows.

Nvidia and Foxconn Push Agentic AI Into Taiwan Hospitals

Image source: 인공지능신문

What happened

Nvidia and Foxconn are moving agentic AI into Taiwan medical institutions after an announcement at Nvidia GTC Taipei on June 1 local time.

The effort links Nvidia, Foxconn and major Taiwanese medical centers with the Healthy Taiwan initiative, which includes USD 1.5 billion for a sovereign AI healthcare ecosystem.

The project is important because it treats healthcare AI as a coordinated operating layer, not only as a diagnostic tool.

The planned systems are meant to analyze patient conditions, review medical records, coordinate treatment plans and support collaboration across hospital departments.

Why it matters

Taiwan faces pressure from rapid population aging and medical workforce shortages.

Hospitals are therefore a practical test case for agentic AI, because adoption will depend on whether systems can reduce workload while keeping clinical safety and accountability intact.

Foxconn is using its CoDoctor platform for specialist AI agents in cardiovascular disease, oncology and ophthalmology.

Named tools include ECG analysis, Corovia for automatic three-dimensional coronary reconstruction and Endovia for real-time lesion detection during colonoscopy.

Corovia can shorten cardiovascular analysis from about two hours to less than one minute.

Foxconn also introduced CoDoClaw, a clinical AI platform based on Nvidia NemoClaw.

It manages agents for breast cancer screening, ECG analysis, fundus image reading and coronary artery analysis, and is already used by major Taiwanese medical institutions for heart disease diagnosis, cancer classification, surgery planning and colonoscopy support.

What to watch next

The affected market includes hospitals, clinicians and healthcare technology suppliers trying to connect software agents with physical work.

Foxconn showed Scrub Bot for operating rooms and is expanding Nurabot for nursing logistics.

Nurabot can help nurses free an average of two to three hours a day for patient care.

Nvidia and Foxconn are also using Nvidia Omniverse-based hospital digital twins to train and verify AI and robot systems before deployment.

The companies said the method can reduce robot deployment time by up to 40 percent and reach 98 percent navigation accuracy.

The next question is whether the multi-agent model moves from controlled pilots into daily hospital operations.

#ai #Nvidia #Foxconn #agentic AI

Perplexity Makes AI Efficiency the Next Test for Agentic Platforms

Perplexity CEO Aravind Srinivas is positioning AI efficiency around the metric of token value per watt per user. The company's Personal Computer product is an orchestration layer that decides which model to use, how agents cooperate and where AI processing should happen. The market test is whether Perplexity can convert its neutral, cross-model approach into durable value while larger platform companies build their own AI agents.

Microsoft Uses Build 2026 to Push Agents Beyond Copilot

Microsoft used its Build 2026 keynote to introduce MAI models, Project Soltera and Microsoft Scout as part of a broader agent strategy. MAI-Thinking-1 is described as a 35-billion-parameter reasoning model with a 128,000-context window for multi-step instructions, long-context reasoning and code generation. The announcement gives Microsoft a clearer agent roadmap, but the source does not provide customer rollout data, pricing or enterprise adoption evidence.

AI Token Costs Push Enterprises Toward a New Spend-Control Layer

Companies are moving from broad AI adoption to stricter control of token spending as agentic tools raise internal usage and budget pressure. The Linux Foundation unveiled plans for the Tokenomics Foundation, while Faros and Jellyfish data point to higher developer output alongside bugs, rewrites and sharply higher token consumption. The next signal is whether common token standards and spend-management tools can give enterprises enough visibility before AI budgets tighten further.

Coralogix's $200 Million Round Puts AI-Agent Monitoring on the Enterprise Watchlist

Coralogix raised $200 million in Series F financing to expand software-monitoring tools for AI-agent operations. The round valued the company at $1.6 billion post-money and brought total capital raised to $550 million. The practical test is whether enterprise use of AI agents turns observability spending into durable growth for Coralogix.