Nikkei Surges 2.53% to 70,037.61, Chip Stocks Lead

Key Takeaways

• <ul><li>Japan's Nikkei hit a 3-month high above 70,000, driven by AI and chip stock gains</li><li>Teradyne invested in Bright Machines to speed up AI infrastructure manufacturing across its 130+ microfactories</li><li>OpenAI leads enterprise AI agent security adoption at 40%, with Microsoft Azure at 22%</li><li>97% of AI-linked malware never reached production, according to Palo Alto Networks research</li><li>Zeta Global is acquiring Senso to improve brand discoverability across ChatGPT, Gemini, and Claude</li><li>Corpay expanded AI agents on its platform despite a 5.23% stock decline over 30 days</li><li>Dynatrace acquired Arize to strengthen AI observability for developers</li><li>CData launched Connect AI Gateway to help manage AI costs and governance</li><li>Jiuzi Holdings expects $1.55M initial profit from its AI Smart Imaging Platform</li><li>New CISE method improves reliability of self-evolving AI search in materials science</li></ul>

Japan's Nikkei hits 3-month high above 70000 on AI stock rally

Japan's Nikkei rose 2.53% to 70037.61 on Monday, crossing 70000 for the first time in three months. The broader Topix also gained 1.16% to 4138.57. Chip stocks led the gains, with Advantest and Tokyo Electron climbing 4.5% and 5.7%. The rally followed Wall Street gains and weaker US jobs data that reduced expectations for a Federal Reserve rate hike.

Japan's Nikkei reaches 3-month peak as AI and chip stocks surge

Japan's Nikkei surged 2.53% to 70037.61, its highest level since early July. AI and chip-related stocks drove the gains, led by Advantest and Tokyo Electron. SoftBank Group also rose 3.8%. Investors remained cautious about the AI rally, with strategist Mamoru Shimode saying markets would become more selective. Easing expectations for a Federal Reserve rate hike supported sentiment.

Teradyne invests in Bright Machines to boost AI manufacturing

Teradyne made a strategic investment in Bright Machines to advance AI infrastructure manufacturing. The partnership combines Teradyne's robotics and test technologies with Bright Machines' software-defined manufacturing platform. The companies plan to explore integrating precision robotic assembly and autonomous material movement into Bright Machines' factories. Bright Machines has deployed over 130 microfactories across more than 10 countries.

Teradyne invests in Bright Machines for AI infrastructure production

Teradyne and Bright Machines announced a strategic investment and collaboration on October 5, 2026. The deal integrates Teradyne's robotics and test equipment with Bright Machines' software-defined manufacturing platform. The goal is to help companies build AI infrastructure faster by connecting assembly, robot, and test data streams. Bright Machines CEO said the partnership supports a more automated and data-driven approach to manufacturing.

Corpay expands AI agents as valuation debate continues

Corpay added new AI agents to its Corpay Complete platform for spend analysis, coding automation, and voice-based expense handling. The stock fell 5.23% over the past 30 days but rose 10.50% over 90 days and gained 36.68% over the past year. Analysts see fair value at $461 versus a recent close of $394.60. Corpay trades at a 22.9x P/E ratio compared to a fair ratio of 16.8x.

Dynatrace acquires Arize to strengthen AI observability

Dynatrace acquired Arize to improve how developers monitor AI applications from development through production. Arize co-founders Jason Lopatecki and Aparna Dhinakaran will join Dynatrace. Arize will continue supporting its Phoenix open-source offering and AX enterprise product while its capabilities are gradually integrated into Dynatrace's platform.

OpenAI leads enterprise AI agent security adoption at 40%

A VentureBeat Intelligence survey found that 40% of organizations with 100 or more employees use OpenAI's built-in guardrails as their primary AI agent security layer, up from 21% in July. Microsoft Azure followed at 22%. Organizations relying on OpenAI were nearly three times as likely to say attackers using AI are ahead of their defenses. Among organizations running agents, 59% reported a security incident or near-miss in the past 12 months.

Study finds 97% of AI-linked malware never reached production

Palo Alto Networks Unit 42 researchers analyzed 405 malware samples linked to artificial intelligence and found only 12, about 3%, were observed on production endpoints protected by Cortex XDR. The remaining 97% were confined to research repositories, sandboxes and testing environments. Every operational sample was detected using existing conventional malware technologies, meaning AI changed how code is authored but not how it executes.

Zeta Global to acquire Senso for AI engineering and model connectivity

Zeta Global signed a definitive agreement to acquire Senso, a Y Combinator-backed AI engineering company. The deal adds AI engineering talent and deepens Zeta's connection to frontier models. Senso will help Zeta improve brand discoverability across AI platforms including ChatGPT, Gemini, Claude and Google AI Overviews, and connect those insights to customer intelligence and marketing activation.

CData launches Connect AI Gateway for better AI ROI and governance

CData expanded its Connect AI platform with the Connect AI Gateway, designed to help organizations manage high AI costs, unclear return on investment and governance challenges. The gateway acts as a control point between AI agents and enterprise IT systems, extending the existing CData Connect AI MCP server. It provides improved visibility and governance for securely connecting AI to corporate systems and sensitive data.

Jiuzi Holdings advances AI smart imaging platform toward launch

Jiuzi Holdings is preparing its AI Smart Imaging Platform for enterprise use and expects an initial profit contribution of about US$1.55 million. The company aims to build a unified platform for data collection, algorithm processing, and decision support. A technical cooperation framework was established in May 2026, and system validation advanced in September 2026. Commercialization remains in preparation and the profit target still faces testing.

New method CISE improves reliability of self-evolving AI search

Large language model self-evolving search can aid scientific discovery but often uses cheap and imperfect proxy rewards. These flawed rewards can produce false positives that contaminate results and future feedback. The proposed CISE method uses conformal inference to create reward intervals and only returns candidates that pass strict feasibility checks. Tests in materials science showed all CISE results were true positives under high-fidelity evaluation.

Sources

• Japan's Nikkei jumps 2.5% above 70,000 to three-month high as AI stocks rally
• Global Market: Japan’s Nikkei jumps 2.5% to 3-month high as AI stocks rally
• Teradyne Makes Strategic Investment in Bright Machines to Advance AI Infrastructure Manufacturing
• Teradyne Makes Strategic Investment in Bright Machines to Advance AI Infrastructure Manufacturing
• Corpay (CPAY) Expands AI Agents As Undervalued Narrative Faces A Fresh Test
• Dynatrace Acquires Arize to Expand AI Observability
• OpenAI leads AI agent security at 40%
• Study Finds 97% of AI-Linked Malware Samples Never Reached Production Systems
• ChatGPT, Gemini and Claude shape brand discovery. Zeta agreed to buy Senso.
• CData launches Connect AI Gateway to improve AI ROI and governance
• One AI platform could serve stores, security and remote inspections
• Reliable Self-Evolution with Imperfect Proxy Rewards