Generative AI and agentic AI tools are only as good as the problem that they are used to solve. In some cases, using generic AI tools can help with non-specific issues. However, Raj Shukla, of enterprise AI platform vendor Symphony AI, says the future of AI technology will focus on vertical applications and open models. In this Targeting AI episode from AI Business, he emphasizes that open source models provide flexibility and the ability to fine-tune for specific use cases.

Featuring: Raj Shukla, CTO, Symphony AI

In this episode, we cover:

• Symphony's AI mission of bringing AI technology to legacy industries that may struggle with adoption.
• A vertical approach combines predictive, generative and agentic AI to address specific challenges.
• The move in vertical areas from a traditional rule-based approach to a more dynamic, non-deterministic tool.
• AI applications in these verticals can significantly improve operational efficiencies and strategic decision-making.

To learn more about vertical AI applications, check out AI Business from Informa TechTarget, and please subscribe to our newsletter to keep up to date on the most important AI news.

To watch video clips from our podcast, subscribe to our YouTube channel, @EyeonTech.

References:

Small Language Models Gaining Ground at Enterprises

Vertical AI agents explained: The future of enterprise tech

AI21 releases open source tiny language model

President Donald Trump signed an executive order last week that looks to override AI state laws in favor of a national policy. Titled "Ensuring a National Policy Framework for Artificial Intelligence," it directs the Department of Justice to establish an AI Litigation Task Force and challenge "cumbersome" state laws. It also asks the Secretary of Commerce to consider withholding federal funds from states found to have restrictive AI laws. In this podcast, Michael Bennett discusses what the EO means for states like New York and California, which already have established laws in place, and how they might respond.

Featuring: Michael Bennett, Associate Vice Chancellor for Data Science and Artificial Intelligence Strategy, University of Illinois Chicago

In this episode, we cover how:

• The EO aims to prevent conflicting state laws on AI.

• States with existing AI regulations are likely prepared to resist the EO.

• The U.S. has a more laissez-faire approach to AI regulation compared with the EU and China.

• The order could lead to significant political battles leading up to the midterm elections.

• The effectiveness of minimal regulation in winning the AI race is uncertain.

To learn more about AI regulations, check out AI Business, and please subscribe to our newsletter to keep up to date on the most important AI news.

To watch video clips from our podcast, subscribe to our YouTube channel, @EyeonTech.

References:

• Navigating Big Tech’s Influence on the AI Regulatory Landscape in 2025

• Big Tech Firms Ask for AI Regulation but Quietly Hedge Their Bets

• US State Attorneys General Demand Greater AI Safety From Tech Giants

In this episode of the Targeting AI podcast from AI Business, host Esther Shittu interviews Oren Michels, of 2024 startup Barndoor.ai, an AI data and access management vendor, about how to effectively secure enterprise agentic and generative AI systems. The approach is different from traditional cybersecurity paradigms designed to prevent outside intruders from doing harm within an organization's IT system, according to Michels. With agents, security procedures need to focus on the agents themselves to ensure they are performing as their human counterparts intend. The podcast was recorded at the AI Summit conference in New York City on Dec. 10.

Featuring Oren Michels, founder and CEO of Barndoor.ai

In this episode, we cover:

• How enterprises can secure agentic and generative AI systems.
• What mistakes businesses make that make them vulnerable to security threats to AI systems.
• Some of the biggest security threats to large-scale business users of generative and agentic AI technology.
• How to use the Model Context Protocol standard with cybersecurity measures to protect and govern AI agents.

To learn more about security for generative and agentic AI systems, check out AI Business, and please subscribe to our newsletter to keep up to date on the most important AI news.

To watch video clips from our podcast, subscribe to our YouTube channel, @EyeonTech.

References:

• AI Agent Security: Whose Responsibility Is It?
• Governance Is Top Priority for Companies Using Agentic AI: Survey
• What Agentic AI Means for Cybersecurity