AI

Companies Have a New AI Problem: Too Many Agents

By / Blog / , ,

As AI agent adoption grows rapidly in businesses, companies like Lyft, DaVita, and GitLab are facing challenges with “AI agent sprawl,” where too many independently created AI bots complicate cybersecurity, management, and costs. While AI agents improve productivity by automating tasks, firms are now implementing governance and centralized controls to manage proliferation and ensure financial and operational responsibility.

https://www.wsj.com/cio-journal/companies-have-a-new-ai-problem-too-many-agents-9539c4d6

I Don’t Think AI Will Make Your Processes Go Faster

By / Blog / , , ,

The article argues that AI will not inherently speed up processes, especially in software development, because the main bottleneck is often unclear or incomplete problem definitions rather than execution speed. It emphasizes that improving process throughput requires focusing upstream on providing clear, detailed information and predictable inputs to bottlenecks, rather than simply adding resources or relying on AI-generated solutions.

https://frederickvanbrabant.com/blog/2026-05-15-i-dont-think-ai-will-make-your-processes-go-faster/

Every AI Subscription Is a Ticking Time Bomb for Enterprise

By / Blog / , ,

AI providers like OpenAI, Anthropic, and Google are currently heavily subsidizing enterprise AI subscriptions, offering services at prices far below their actual operational costs. However, as advanced agentic AI usage rapidly increases computational demands, these companies face unsustainable losses and will soon need to raise prices or shift to usage-based billing models, posing significant financial risks for enterprises that have integrated AI deeply into their workflows without tracking real consumption costs.

https://www.thestateofbrand.com/news/ai-subscription-time-bomb

Reflections on Science History: a Professor’s Take on AI

By / Blog / , ,

Associate Professor of History David Hecht reflects on the parallels between the atomic age and the rise of artificial intelligence (AI), emphasizing that technological advancements are shaped by social, political, and cultural factors rather than occurring inevitably. Hecht highlights the importance of understanding the societal context that fosters technology, warns against relying solely on fear to shape AI policy, and calls for articulating positive visions for regulating AI to ensure beneficial outcomes.

https://bowdoinorient.com/2026/05/16/reflections-on-science-history-a-professors-take-on-ai/

Anthropics/claude-For-Legal: a Suite of Plugins for Legal Workflows

By / Blog / , , ,

The “claude-for-legal” GitHub repository by Anthropics offers a comprehensive suite of AI-powered plugins, agents, and connectors designed to support a wide range of legal workflows—including commercial, corporate, employment, privacy, litigation, IP, and AI governance. These tools integrate with Claude Cowork or Claude Code and include practice-area-specific skills, scheduled agents, and research connectors, facilitating efficient legal review and drafting while emphasizing that all outputs are drafts requiring attorney review and responsibility.

https://github.com/anthropics/claude-for-legal

How Deepfakes Are Rewriting the Rules of the Modern Workplace

By / Blog / , ,

Deepfake technology is increasingly impacting the modern workplace by enabling sophisticated impersonation attacks that exploit trust in familiar voices and faces, leading to significant security risks such as fraudulent payment approvals and misinformation. Organizations must adapt by implementing stricter verification processes, expanding incident response plans to address synthetic media threats, and applying zero-trust principles to communication channels to safeguard against these evolving digital manipulations.

https://www.cio.com/article/4170894/how-deepfakes-are-rewriting-the-rules-of-the-modern-workplace.html

The Shadow AI Jungle: Why Approving a Platform Is Not the Same as Securing What’s Built on It

By / Blog / ,

The article highlights a critical security concern in enterprise AI adoption dubbed “Shadow AI,” where non-technical employees build AI tools and automations on approved platforms without security oversight, creating significant blind spots for security teams who can track less than half of these AI agents. Despite platform approvals, enterprises remain responsible for securing what is built on them, yet many AI tools operate invisibly, often accessing sensitive data without triggering alerts, underscoring the urgent need for runtime governance and visibility into these business-built AI applications to manage risks effectively.

https://www.unite.ai/the-shadow-ai-governance-challenge/

The Death of Identity as We Know It

By / Blog / ,

In “The death of identity as we know it,” Steve Tout discusses the evolving challenges of AI governance, emphasizing that identity must shift from traditional authentication toward authorship and lineage of AI entities like agents, swarms, and digital twins. He highlights the necessity of new governance models that track who creates, trains, authorizes, and controls AI-powered digital representations to ensure accountability, protect institutional knowledge, and prevent misuse as AI becomes integral to enterprise decision-making.

https://www.cio.com/article/4170235/the-death-of-identity-as-we-know-it.html

Culture Is Critical for AI Project Success

By / Blog / ,

A Microsoft report finds that organizational readiness, including a supportive culture, clear policies, and managerial backing, is the leading factor for successful AI pilot projects, yet only about 20% of employees currently operate with both high individual AI skills and effective organizational infrastructure. Experts emphasize that companies must redesign workflows, foster AI experimentation, and build robust infrastructure and governance to enable widespread AI adoption and sustainable results.

https://www.ciodive.com/news/culture-critical-for-ai-success/819902/

Software Bill of Materials for AI – Minimum Elements

By / Blog / , ,

The Cybersecurity and Infrastructure Security Agency (CISA) outlines the minimum elements for a Software Bill of Materials (SBOM) specific to AI systems to enhance transparency and security. These elements include detailed information about the components, versions, and relationships within AI software to help identify vulnerabilities and manage risks effectively. This approach aims to improve trust and security in AI technologies by providing comprehensive visibility into their software components.

https://www.cisa.gov/resources-tools/resources/software-bill-materials-ai-minimum-elements

Scroll to Top