42Gears says its SureMDM platform has earned ISO/IEC 42001:2023 certification, the international standard for AI management systems. The British Standards Institution independently audited its governance, risk management and human-oversight frameworks, including safeguards for third-party large language models. The certification covers 42Gears as a producer, provider and user of AI and complements its ISO/IEC 27001, SOC 2 Type II, PCI DSS and HIPAA programs. CEO and co-founder Onkar Singh said it supports responsible AI use in areas including banking, healthcare and critical operations. The company says it serves more than 23,000 customers in over 170 countries through more than 1,000 partners.
Two security articles warn that AI agents can create serious exposure even when acting with valid credentials and permitted access. One describes an accounts employee’s email-and-ERP agent, built with OAuth grants and a personal API key without the security team’s knowledge. It could read untrusted email, access private vendor banking details and change ERP records. The guidance is to review agents before launch, map their identity and permissions, restrict access to task-specific, short-lived credentials, monitor activity and require human approval for high-impact actions. The articles cite CrowdStrike’s finding of an 89% increase in attacks by AI-enabled adversaries in 2025 versus 2024, and an Anthropic report estimating that a state-sponsored group used AI for 80% to 90% of tactical operations in an espionage campaign.
AI is also speeding up course development while educators retain responsibility for quality. A university team says drafting a video-lesson script now takes half a day rather than at least seven working days, while a course map can be drafted in under 30 minutes instead of at least three weeks. Instructional designers review the learning design, and academic staff check accuracy and relevance.
At a November 9 New York event, advertising executives said automation is shifting competition toward the quality of signals, creative choices and ad-placement decisions. AI assistants may combine product research, discovery and comparison, although consumers still participate in decisions. Scale, decision-making intelligence and trust could shape competition and consolidation. Ravi Patel of SWYM.ai said unified platforms can manage budgets across Google, Meta and Amazon, but cannot select individual impressions within those companies’ systems.
Other business coverage includes an investment discussion with Prime Capital Financial CIO Will McGough, who says investors are looking beyond the AI trade even as AI may have further room to grow. Intuit is promoting AI for tax preparation and bookkeeping, while SES AI announced third-quarter 2026 results on November 12; the supplied summaries provide no further financial details for either company. Modine shares rose about 35% Friday morning, then stood 28% higher, after reports that Samsung Electronics was interested in acquiring the heating-and-cooling systems supplier.
Zoom has backed Medaica’s Telehealth 2.0 platform and connected Quiq’s agentic AI with Zoom Contact Center. Its shares were reported at US$97.70, up about 10% over seven days but down roughly 7.8% over 90 days. Zoom’s one-year total shareholder return was 16.25%, while its five-year return was a loss of 61.8%. Analysts expect annual revenue growth of 4% over three years and earnings of $1.3 billion by about October 2029, compared with $3.3 billion today. They also flag bundled-suite competition and rising AI computing costs as potential limits on pricing power and margins; shares were about 18% below analyst targets and 34% below estimated intrinsic value.
Cybersecurity researchers say AI has compressed complex attack chains from weeks or months to about 20 minutes. Mitos demonstrated the speed, and Luttwak said models can find and exploit complex vulnerabilities; open models are reportedly about three months behind frontier models. Slow patching remains a weakness: about half of Wiz customers patched PHP within two to three days, despite PHP powering about 70% of the web. Wiz now continuously tests frontier models against critical code, and multiple teams have reported the same critical flaw in its Zero Day competition.
Key Takeaways
- 42Gears earned ISO/IEC 42001:2023 certification for AI management in SureMDM after an independent British Standards Institution audit; it covers governance, risk, human oversight and third-party large language models.
- 42Gears says it serves more than 23,000 customers in over 170 countries through more than 1,000 partners.
- AI agents can combine access to untrusted inputs, private information and consequential actions; security guidance calls for pre-launch review, least-privilege credentials, monitoring and human approval for high-impact changes.
- CrowdStrike reported an 89% rise in attacks by AI-enabled adversaries in 2025 versus 2024; an Anthropic report estimated AI handled 80% to 90% of tactical operations in one state-sponsored espionage campaign.
- A university team cut video-lesson script drafting from at least seven working days to half a day, and course-map drafting from at least three weeks to under 30 minutes, with staff checking design and accuracy.
- Advertising competition is shifting toward useful signals and decisions; unified platforms may manage budgets across Google, Meta and Amazon but do not choose individual impressions within those systems.
- Zoom has backed Medaica’s Telehealth 2.0 and integrated Quiq agentic AI with Zoom Contact Center; analysts cite competition and AI costs as risks to margins.
- Analysts expect Zoom earnings of $1.3 billion by about October 2029, down from $3.3 billion today, and annual revenue growth of 4% over three years.
- Reports of Samsung Electronics’ interest in Modine preceded a roughly 35% initial rise in the supplier’s shares, which were later up 28%.
- Mitos demonstrated an AI attack chain in about 20 minutes; PHP patching remains uneven, with about half of Wiz customers patching within two to three days despite PHP powering about 70% of the web.
42Gears earns ISO certification for AI management systems
42Gears announced on November 13, 2026, that it earned ISO/IEC 42001:2023 certification for its artificial intelligence management systems. The certification covers AI systems in its SureMDM platform and the company’s roles as a producer, provider, and user of AI. It recognizes controls for AI governance, risk assessment, transparency, human oversight, and third-party large language models. The certification adds to 42Gears’ existing ISO/IEC 27001, SOC 2 Type II, PCI DSS, and HIPAA programs.
42Gears earns global certification for AI oversight in SureMDM
42Gears earned ISO/IEC 42001:2023 certification, the international standard for Artificial Intelligence Management Systems. An independent audit by the British Standards Institution validated the AI governance, risk management, and human oversight frameworks in its SureMDM platform. The certification covers 42Gears’ work as a producer, provider, and user of AI systems. It also addresses risk assessment, transparency, third-party large language models, and compliance alongside ISO/IEC 27001, SOC 2 Type II, PCI DSS, and HIPAA programs.
42Gears certification backs AI controls in SureMDM
42Gears announced on November 13, 2026, that it earned ISO/IEC 42001:2023 certification for the AI systems in its SureMDM platform. The British Standards Institution conducted an independent audit of the company’s AI governance, risk management, and human oversight. The certification covers 42Gears as a producer, provider, and user of AI, and includes safeguards for third-party large language models. CEO and co-founder Onkar Singh said the certification supports responsible AI use in fields such as banking, healthcare, and critical operations. 42Gears says it serves more than 23,000 customers in over 170 countries through more than 1,000 partners.
Cloud controls help limit risks from AI agents
AI agents can cause security incidents even when they use valid credentials and take permitted actions, such as exposing a private database while restoring a cloud service. The article says overly broad permissions, functionality, or autonomy can increase these risks. It cites an 89% rise in attacks by AI-enabled adversaries in 2025 compared with 2024, according to CrowdStrike, and an Anthropic report that a state-sponsored group used AI for an estimated 80% to 90% of tactical operations in an espionage campaign. It urges companies to check software and deployments, give agents short-lived and task-specific access, and monitor their actions, with human approval for high-impact changes.
An everyday AI agent reveals five enterprise security gaps
The article follows an AI agent built by an accounts employee to read email, match invoices with an ERP system, and post summaries to Slack. Its creator gave it access through email and Slack OAuth grants and a personal ERP API key, while the security team did not know the agent existed. The article explains that agents may use a person’s session, delegated OAuth, static credentials, or purpose-built agent identities, and each method affects what appears in security logs. It warns that this agent could read untrusted email, view private vendor bank details, and change ERP records, combining three risky abilities. It recommends reviewing such agents before launch and involving security when they can access outside input, private data, and take action without human review.
Investors look beyond the AI trade for new opportunities
Will McGough, CIO of Prime Capital Financial, discusses investment opportunities beyond the AI trade in a video. He says AI may still have room to grow, but investors are increasingly asking how to diversify away from it. McGough outlines other investment opportunities.
AI speeds up course design while humans check the work
A university team explains how it combines AI tools with human judgment to build competency-based courses. AI has cut the time to draft a video lesson script from at least seven working days to half a day, and a course map can now be drafted in under 30 minutes instead of taking at least three weeks. In an engineering bachelor’s programme, the team used existing course materials and new learning outcomes to create draft reports and course maps. Instructional designers check the learning design, while academic staff validate subject matter accuracy and relevance.
AI shifts ad tech competition toward smarter decisions
AI is making advertising workflows easier to automate, shifting competition toward companies that can choose useful signals, creative, and ad placements to improve results. At a November 9 event in New York City, speakers said AI could intensify competition among advertising platforms, agencies, and retail media networks. AI assistants may also combine product research, discovery, and comparison, while consumers remain involved in decisions for now. Speakers said scale, useful signals, decision-making intelligence, and consumer trust will matter, and that the changes could lead to a more competitive market and consolidation. Ravi Patel of SWYM.ai noted that unified media platforms can manage budgets across platforms but cannot choose individual impressions, which remain within systems such as Google, Meta, and Amazon.
Intuit uses AI to simplify personal finances
Intuit is using AI to make tax preparation smarter and bookkeeping more automated. Peloton instructor and former CPA Matt Wilpers joins the discussion to explore how technology is changing the financial landscape.
SES AI reports third quarter 2026 financial results
SES AI Corporation reported its third quarter 2026 financial results. The announcement came from Woburn, Massachusetts, on November 12, 2026.
Zoom’s AI and telehealth moves meet mixed growth outlook
Zoom Communications has backed Medaica’s Telehealth 2.0 platform and connected Quiq’s agentic AI with Zoom Contact Center. The company’s shares traded at US$97.70, up about 10% over seven days but down roughly 7.8% over 90 days. Its one year total shareholder return was 16.25%, while its five year return was a loss of 61.8%. Zoom trades about 18% below analyst targets and 34% below estimated intrinsic value, while analysts expect annual revenue growth of 4% over three years and earnings of $1.3 billion by about October 2029, down from $3.3 billion today. Analysts warn that competition from bundled suites and higher AI computing costs could limit pricing power and reduce margins.
Samsung reportedly considers buying data center supplier Modine
Modine Manufacturing shares surged about 35% Friday morning after reports that Samsung Electronics was interested in buying the company. After the initial rise, shares were up 28%, according to MarketSurge. Modine makes heating and cooling systems for industrial buildings. Its business has boomed in the past couple of years, though the article does not provide further details.
AI shortens complex cyberattacks from months to 20 minutes
Mitos showed that AI can carry out an attack chain in about 20 minutes, compared with weeks or months for the best attackers before. Luttwak said models can now find and exploit complex vulnerabilities, and open models have narrowed the gap to about three months. This may let less skilled attackers exploit API mistakes and PHP flaws. Patching remains slow: about 50% of Wiz customers patched PHP within two to three days, even though PHP powers about 70% of the web. Wiz now tests frontier models continuously against critical code, and multiple teams have reported the same critical flaw in its Zero Day competition.
Sources
- 42Gears Achieves ISO/IEC 42001:2023 Certification for Artificial Intelligence Management Systems
- 42Gears Achieves ISO/IEC 42001:2023 Certification for Artificial Intelligence Management Systems
- 42Gears Achieves ISO/IEC 42001:2023 Certification for Artificial Intelligence Management Systems
- When AI Agents Have Cloud Access, Who Has Control?
- How one AI agent exposed five critical enterprise security gaps
- Opportunities beyond the AI trade
- Can AI design your course?
- AI is turning ad tech into a fight for the brain behind the buy
- Intuit
- SES AI Reports Third Quarter 2026 Financial Results
- Zoom Communications (ZM) Could Be 17% Undervalued As AI And Telehealth Moves Build
- Samsung Eyeing Deal To Buy Data Center Supplier Modine; Stock Soars 35%
- AI cut attacks that took months to 20 minutes
Comments
Please log in to post a comment.