AI Experts Say New Path Needed for Human-Like Intelligence

In the rapidly evolving world of artificial intelligence, the latest advancements continue to capture global attention. Yet, amidst this whirlwind of innovation, a survey conducted by the Association for the Advancement of Artificial Intelligence (AAAI) casts a shadow of doubt over the path to mimicking human intelligence. Could we be chasing an elusive dream?

Survey Reveals Growing Skepticism

The AAAI surveyed 475 AI specialists worldwide, uncovering that a whopping 76% believe that current models are unlikely to achieve artificial general intelligence (AGI). This skepticism suggests an urgent need to re-evaluate the trajectory of AI development. According to EL PAÍS English, experts are questioning whether pursuing the improvement of existing models will suffice in reaching AGI.

A Sector Divided

While there are voices, like the group AI 2027, predicting a future dominated by superhuman AI, others stay grounded. Figures such as José Hernández-Orallo from the Leverhume Centre for the Future of Intelligence express hope in current AI technologies, albeit with caveats regarding their scalability and resource demands.

Meta’s Recent AI Models Under Scrutiny

Amidst this debate, companies continue to push boundaries, with Meta recently introducing Llama 4 and Maverick models. Surprisingly, the latter captured second place in competitive rankings. However, investigations revealed Maverick’s tailored training for specific tests, raising ethical questions within the field.

Defining Human Intelligence: The Current Challenge

Experts emphasize the complexity of defining and replicating human intelligence. Nuria Oliver, of the Ellis Alicante Foundation, highlights the distant reality of achieving self-awareness in AI. Meanwhile, Senén Barro from the University of Santiago de Compostela compares the journey to AGI to human travel to Mars—full of challenges, but ultimately attainable with breakthroughs in technology.

Physical and Cognitive Dimensions in Focus

Further complicating the landscape, researchers such as Carme Torras advocate for AI to embody corporeal elements and reasoning capabilities. This holistic approach aligns with Carles Sierra’s vision of integrating neurosymbolic methodologies to foster AGI.

As the debate rages on, the future of AI seems both hopeful and uncertain, demanding a collective reassessment of strategies to truly emulate the rich tapestry of human thought.

Join us in exploring diverse AI insights by subscribing to our newsletter. Let’s navigate this fascinating journey together, keeping you informed and engaged with step-by-step developments in the AI landscape.