What factors are most important for driving the adoption of AI agents?

Most answers emphasize trust, confidence, and human involvement as key to AI agent adoption, while others highlight technical factors like data access, reliability, and clear value.

12 answers

Answers

  • Chris Lattner

    EVP of Advanced AI Software and Platforms · Qualcomm

    Experimenting with Tools

    Chris Lattner says people need to be experimenting with these tools and leaning into AI, because the risk is falling behind rather than being obsoleted.

    The TWIML AI Podcast with Sam Charrington · Watch on YouTube
  • David Vellante

    Co-founder, co-CEO SiliconANGLE, theCUBE, theCUBE Research (formerly Wikibon) · Siliconangle & Thecube

    Networking

    David Vellante says networking is the common thread in driving AI agent adoption, because it makes data move faster and that enablement changes the game.

    The answer starts at 31:42.

    theCUBE Podcast · Listen on Apple Podcasts
  • Joe Albrecht

    Wealth Strategy Associate · Ubs

    Human Involvement

    Joe Albrecht says the limiting factor in AI adoption is the human side rather than technology, calling it a people problem.

    Imagination in Action · Watch on YouTube
  • Prabhav Jain

    CEO · 11x.Ai

    Clear Value

    Prabhav Jain says the best way to drive adoption is to make agents so valuable that people pull them from you, since it's hard to force people to use something they see no value in.

    FinStrat Management, Inc. · Watch on YouTube
  • Rama Akkiraju

    VP, Enterprise AI · Nvidia

    Observability

    Rama Akkiraju says observability is key, especially for understanding agent behavior before an agent can be certified.

    Imagination in Action · Watch on YouTube
  • Jonathan Evens

    Product Lead · Google Deepmind

    Trust

    Jonathan Evens says trust is the key factor, explaining that for products like Google search it means factuality and grounding in trusted sources, plus a user experience that shows citations so users can quickly judge whether to trust the information.

    Mind the Product · Watch on YouTube
  • Dilani Kahawala

    Co-Founder and CEO

    Reliability, Interface, Teaching Users

    Dilani Kahawala says reliability, the interface, and teaching users how to work with an agent have been the three biggest challenges her team has had to think about.

    Super Data Science: ML & AI Podcast with Jon Krohn · Watch on YouTube
  • Zeus Kerravala

    Founder and Principal Analyst · Zk Research

    Context Across Channels

    Zeus Kerravala says it is critically important that context exists across channels, different tools and systems, and that human-digital agent handoff works, because fragmented data and stale knowledge limit accuracy and frustrate customers.

    SiliconANGLE theCUBE · Watch on YouTube
  • Ben Lorica 罗瑞卡

    Principal · Gradient Flow

    Confidence in the Agent

    Ben Lorica says teams should keep humans involved and reduce that involvement over time as they develop confidence in the agent, and argues it's not good practice to let an agent run wild right now.

    The answer starts at 15:36.

    The Data Exchange with Ben Lorica · Listen on Apple Podcasts
  • Tara Seshan

    Member of Product Staff · Openai

    Data Access & Infrastructure

    Tara Seshan says a huge part of making agents useful is tactical work like data access, cloud infrastructure and reliability, which she says matters just as much as intelligence for end effectiveness.

    Lenny's Podcast · Watch on YouTube
  • Peter Deng

    General Partner · Felicis Ventures

    Trust

    Peter Deng says trust is the key distinction between AI agents, noting people are only starting to trust AI more and that fear persists until people really experience it.

    More or Less Podcast · Watch on YouTube
  • Katie Leland

    Chief of Staff to CVP | Global Customer Success | Microsoft Customer and Partner Solutions · Microsoft

    Trust and Confidence

    Katie Leland says success means giving users the trust and confidence to use AI for their needs, arguing that guardrail policies should remove the end user's need to second-guess their AI usage.

    N2K Networks · Watch on YouTube