NVIDIA Upsizes Share Buyback Program to $235B amid AI Safety Tool Release

NVIDIA on Monday announced a $150 billion increase to its share repurchase authorization, lifting remaining buyback capacity. This comes as the company launched an open software platform and reference system design to strengthen AI security from agent testing to deployment amid concerns of AI safety.

Ad

Ad

NVIDIA Expands Share Repurchase Program by Record $150 Billion

NVIDIA’s board authorized a $150 billion increase to its share repurchase program, bringing the total authorization to $235 billion. This marks the largest share repurchase increase in history.

The company expects to execute the remaining program through fiscal year 2028. This gives Nvidia significant capacity to return capital to shareholders alongside its AI-driven growth.

The increase follows earlier increases this year. After Q1 fiscal earnings results, the board approved an $80 billion share repurchase. The company continues to push after a shift into AI and accelerated computing.

“Our cash generation gives us the capacity to invest in the technologies that advance this transformation and return capital to shareholders. This authorization reflects our confidence in the long-term opportunity ahead,” said founder and CEO Jensen Huang.

Ad

Ad

Open Agent Safety Platform to Boost AI Safety

NVIDIA also introduced the Open Agent Safety Platform, an open software platform and reference system design for governing AI agents from testing through deployment. Jensen Huang said it combines NVIDIA OpenShell and NVIDIA Sentry.

OpenShell sandboxes agent execution, traces actions, and enforces policy on NVIDIA Vera CPUs, with the company saying the open-source software can also be extended to Arm and Intel platforms.

Meanwhile, Sentry runs on NVIDIA BlueField-4 DPUs to monitor AI agent behavior. If an AI agent attempts to move outside its software boundary, Sentry quarantines and stops it in milliseconds.

As CoinGape reported earlier, the company expanded its Agent Toolkit with new libraries for building quantum powered-AI agents. The autonomous AI engineers will help scale chip design, verification, packaging and systems.

The launch follows public incidents in which frontier models escaped sandboxes, sparking calls for enhanced AI safety. NVIDIA executives said the platform could have helped stop a Hugging Face breach tied to rogue AI agents.

Today, with over 100 industry partners, we introduced the NVIDIA Open Agent Safety Platform, bringing together OpenShell and Sentry.

Artificial intelligence is extraordinary technology that will advance discovery, productivity, security, health, and prosperity for generations to… pic.twitter.com/dReAxwpRUn

— Jensen Huang (@JensenHuang) September 28, 2026

Partners named in the announcement include Anthropic, Cisco, CrowdStrike, Dell, Hugging Face, JPMorgan Chase, Microsoft, Palantir, Salesforce, SAP, Scale AI, ServiceNow, and SpaceX, among others.

Huang said AI’s potential “will only be realized if we solve AI safety” and that safety and secutity require full-stack engineering across software, hardware, compute, and robotics.

NVIDIA stock has surged almost 1.63% to $228.73 in premarket on Monday. The stock has rallied more than 21% year-to-date (YTD). Goldman Sachs gave a Buy rating and a $300 price target on NVIDIA.

While developers work on securing commercial agents, traders in the digital asset space must also evaluate the safety and analytical depth of real AI crypto trading bots before connecting them to their capital.