We're sunsetting PodQuest on 2025-07-28. Thank you for your support!
Export Podcast Subscriptions
People
N
Nufar Gaspar
英特尔设计部门的 AI 无处不在和通用 AI 总监,专注于推动 AI 技术在企业中的应用和发展。
Topics
Nufar Gaspar: 2025 年,几乎每家公司都会推出 AI 代理产品,或声称其运营背后由 AI 代理提供支持。这既有炒作的成分,也有实际的应用场景。微软、Salesforce、谷歌和 OpenAI 等公司都在积极发展 AI 代理技术。垂直领域的 AI 代理公司将比横向平台公司获得更高的投资回报率,因为它们更专注于特定用例,并能为企业提供更直接的价值。 在工作场所,AI 代理将被广泛应用,成为知识工作者的个人助理,并自动化后台任务。企业将进行大量 AI 代理试点项目,以探索其在不同业务流程中的应用。 到 2025 年底或 2026 年,活跃的 AI 代理数量将超过全球人口数量。AI 代理的成本效益、精确度将得到提升,但其定价模式仍不明确。企业需要制定明确的策略,以确定何时、如何以及由谁创建 AI 代理。 从副驾驶模式转变为代理模式,AI 代理将承担更多自主决策和执行任务。然而,最成功的 AI 代理仍然需要人类的监督和预先规划,尤其是在金融、医疗和法律等高风险行业。 人们对 AI 代理处理信用卡等敏感信息的信任度仍然较低。对 AI 代理对现有工作的影响,以及 AI 代理在替代和增强人类方面的作用,需要进一步的讨论和研究。 在伦理方面,需要关注 AI 代理的公平性、隐私性和安全性,并确保其可解释性。AI 代理的负面影响,例如欺诈、网络犯罪和军事用途,也需要引起重视。 NLW: 同意 Nufar Gaspar 的大部分观点,并补充了一些细节和自己的见解。例如,NLW 认为企业将在 2025 年进行大量 AI 代理试点项目,这反映了企业对 AI 技术的信心和对垂直 AI 代理的期待。NLW 还强调了在 AI 代理应用中,人类监督和预先规划的重要性,以及在不同用例中,人类与 AI 代理协作方式的多样性。此外,NLW 也表达了对 AI 代理潜在的伦理风险和社会影响的担忧,并认为需要通过渐进式实验来理解和应对这些挑战。

Deep Dive

Key Insights

Why does Nufar Gaspar believe that nearly every company will showcase agent-driven features by 2025?

Nufar Gaspar predicts that nearly every company will showcase agent-driven features or claim to have an AI agent at work by 2025 due to the increased feasibility and notable releases from major companies like Microsoft, Salesforce, and Google. These companies have already made significant announcements and releases, indicating a trend where AI agents will power operations behind the scenes across hyperscalers, SaaS, and vertical industries.

What is the difference between a vertical agent and a horizontal agent according to Nufar Gaspar?

Vertical agents are specialized for specific use cases or functions within an organization, such as customer support or coding, and are deeply tied to business context. Horizontal agents, on the other hand, offer the ability to build agents for diverse use cases without specialization, like platforms from Microsoft or OpenAI. Nufar believes vertical agents will yield higher ROI in 2025 due to their practicality and specialization.

What does Nufar Gaspar predict about the widespread adoption of AI agents in the workplace by 2025?

Nufar Gaspar predicts that AI agents will become ubiquitous in the workplace by 2025, acting as co-workers, personal assistants, and support functions. They will handle tasks like email management, HR inquiries, and IT support, often operating autonomously in the background. This widespread adoption will be driven by companies piloting agents across various workflows, embedding them deeply into daily operations.

Why does Nufar Gaspar believe there will be more agents than people by 2025?

Nufar Gaspar predicts that the number of active AI agents will surpass the global human population by 2025 due to the ease of creating agents and their widespread use in consumer and work life. Every company will deploy bots for customer support, and most online interactions will be agentic. This proliferation will lead to a future where humans are surrounded by more virtual workers than actual people.

What does Nufar Gaspar predict about the cost, performance, and precision of AI agents by 2025?

Nufar Gaspar predicts that AI agents will become more cost-effective, faster, and more precise by 2025 due to optimized hardware, smarter algorithms, and additional capabilities that allow agents to learn on their own. The ecosystem will mature, and companies will introduce modular solutions, making powerful agents more accessible, efficient, and reliable for organizations.

What does Nufar Gaspar predict about the shift from co-pilot to agent paradigms by 2025?

Nufar Gaspar predicts a shift from co-pilot to agent paradigms by 2025, where early adopters will move from building custom GPTs to creating agents with greater autonomy and execution capabilities. Agents will handle open-ended tasks and automate processes, offering more business value than co-pilots, which primarily assist with strategy and drafting.

Why does Nufar Gaspar believe that the most successful agents in 2025 will still be highly controlled by humans?

Nufar Gaspar believes that the most successful agents in 2025 will still require human oversight and pre-planning to ensure they act within predefined boundaries. Companies will implement rules, guardrails, and fail-safe protocols to prevent unintended consequences, especially in high-stakes industries like finance, healthcare, and legal. This control will be essential to maintain trust and reliability in agent operations.

What does Nufar Gaspar predict about the ethical considerations and responsible development of AI agents by 2025?

Nufar Gaspar predicts that there will be increased focus on ethical considerations and responsible development of AI agents by 2025. Agents will need to be fair, privacy-aware, security-aware, and explainable to avoid biases and ensure transparency. This will require additional governance and policies to address the challenges posed by autonomous agents in various organizational functions.

What does Nufar Gaspar predict about the dark side of AI agents by 2025?

Nufar Gaspar predicts that the dark side of AI agents will grow by 2025, with bad actors exploiting them for fraud, cybercrime, and military purposes. Agents could be used for phishing attacks, social engineering, and autonomous weapons, posing significant challenges for detection and security. This will require vigilance and countermeasures to mitigate the risks associated with malicious use of AI agents.

Shownotes Transcript

Agents are the most important trend in AI heading into the new year. NLW is joined by Nufar Gaspar to count down 25 predictions for AI agents in 2025.

Nufar Gaspar is a seasoned AI expert and leader with vast experience in incubating and growing AI products, verticals and communities. She is the Director of AI Everywhere and Gen AI for Intel Design, and consults and trains organizations and teams on the usage of AI and building AI products and companies.

Brought to you by:

Vanta - Simplify compliance - ⁠⁠⁠⁠⁠⁠⁠https://vanta.com/nlw

The AI Daily Brief helps you understand the most important news and discussions in AI. Subscribe to the podcast version of The AI Daily Brief wherever you listen: https://pod.link/1680633614 Subscribe to the newsletter: https://aidailybrief.beehiiv.com/ Join our Discord: https://bit.ly/aibreakdown