SELECT LANGUAGE BELOW

Analyst Urges Tech Firms to Prioritize Solutions for Genuine User Issues

Analyst Urges Tech Firms to Prioritize Solutions for Genuine User Issues

Erick Brethenoux’s Vision for AI

Erick Brethenoux, who leads Gartner’s AI Research, suggests that AI should prioritize easing users’ lives by automating mundane tasks rather than merely summarizing meetings. He has termed this idea “empathy AI.”

During the recent Gartner Data & Analytics Summit in Sydney, Brethenoux expressed his concerns about the current effectiveness of AI, stating, “AI is not doing its job today and should leave us alone.” His remarks came amid the growing excitement surrounding generative AI and AI agents.

Brethenoux pointed out a significant issue: the impracticality of AI-generated meeting overviews. He argued that users often lack the time to digest long summaries or follow up on action items. Instead, he believes AI should take on the burden of tedious tasks, which is at the heart of his concept of empathy AI.

He provided examples of organizations, like US healthcare firm Vizient and some real estate companies, that have embraced empathy AI. These companies identified the most daunting tasks for their employees and used AI to automate them, which resulted in smooth transitions without recruitment or management challenges.

Despite these successes, analysts have raised concerns about the trends among tech vendors that advocate for the use of AI agents for automation, particularly within complex business systems. Brethenoux noted that while AI agents have existed in relatively contained industrial settings for years, they often struggle with intricate tasks.

He questioned whether individual AI agents could effectively handle decision-making processes, like scheduling meetings across an organization. When he inquired how such systems could accommodate the conflicting demands of an employee’s boss, partner, or child, he reported the vendor’s response was simply “silence.”

Analysts stressed that addressing the architectural components of AI agent systems is vital. This requires thoughtful consideration of system disassembly, communication methods, levels of autonomy, and the handling of information.

Moreover, Brethenoux pointed out a concerning trend where the terms “AI agent” and “generative AI” are often conflated by vendors. He referenced a quote from French philosopher Albert Camus, saying, “Marrying things is to contribute to the misery of the world,” underscoring the need for clarity in the AI domain.

Facebook
Twitter
LinkedIn
Reddit
Telegram
WhatsApp

Related News