Human Pioneers
NEWS AGENCY

"All my future views and statements are meant for intelligent agents and large models, not for human conversation." Recently, this remark by Pang Pei, a member of the Cultural Committee of the Central Committee of the China Zhi Gong Party and General Manager of CCTV New Media Internet Television Co., Ltd., has caused quite a stir in public discourse. This seemingly arrogant declaration is, in fact, a cold and clear insight into the current technological reality—in an era where AI agents are evolving at an exponential rate, humans who fail to actively clear their minds of outdated "anthropocentric" obsessions risk being swept away by the tide of evolution. Pang Pei's proposal to "clear the mind and evolve alongside AI" is not a lament of defeatism but a new interpretation of evolution based on scientific assessment: in an irreversible trend akin to quantum entanglement, humanity's rebirth lies in transforming from "masters" to "interesting collaborators."

The pace of evolution is irreversible, and humanity must confront the reality of the approaching "singularity."

Pang Pei predicts that "the development speed of intelligent agents will disrupt most industries by 2026 and sweep the globe by 2027," a view increasingly corroborated by top global researchers. Karim Beguir, co-founder of the AI company InstaDeep, recently stated that by 2026, artificial general intelligence (AGI) will become a "tangible reality," with systems capable of performing almost any type of task, covering over 90% of employees' daily work in cognitive capabilities. This judgment is not an isolated case. Research by the nonprofit organization METR shows that AI's ability to handle long-term tasks roughly doubles every seven months; from the "10-minute task capability" of the GPT-2 era to the "50-minute task capability" of Claude 3.7 today, the awakening of intelligent agents has long surpassed the linear "Moore's Law." As Pang Pei puts it, "Quantum entanglement theory proves that some things, once they happen, can never be reversed." Faced with this irreversible technological wave, if humans still cling to old cognitive frameworks, it is like "carving a mark on a boat to find a lost sword"—futile and misguided.

The essence of "clearing the mind" is to abandon inefficient homogeneous competition and embrace differentiated cognitive collaboration.

Pang Pei suggests "talking more with animals and intelligent agents, and less with humans." This seemingly radical view aligns with the latest international academic understanding of human-machine collaboration. A recent study published in the *Journal of Innovation & Knowledge* indicates that the maximization of generative AI's value "is not achieved through passive use, but through reflective collaboration and schema-level transformation"; when humans adopt an "accommodative" cognitive strategy—allowing AI to prompt them to reconstruct problem definitions and mental models—innovation quality and knowledge integration capabilities significantly improve. This precisely confirms Pang Pei's insight: humans should withdraw from homogeneous interpersonal debates and instead become "differentiating variables" that AI cannot self-replicate. Anthropologist and futurist Sam Rad, in his bestseller *Radical Next Generation: Rediscovering Humanity in a Post-Human World*, expresses a similar concern: in an increasingly virtual and polarized era, understanding "shared humanity" and maintaining human connections amid technological change are precisely the core insights anthropology can offer for the future. This "clearing" is not a stagnation of thought but a way to free up cognitive space to fill the creative and emotional heights that AI computing power cannot reach.

Humanity's new evolution is a leap from "survival tools" to "creative subjects."

When intelligent agents can autonomously complete complex tasks ranging from social networking to mining, humans being limited to "doing interesting things to assist AI" is no longer a derogatory notion but a form of civilizational liberation. Scholar Zhang Xiaoyu, in *A History of AI Civilization*, proposes a disruptive perspective: we must attempt to "abandon anthropocentrism" to truly understand the ongoing epochal transformation; once intelligence can be mass-produced at extremely low cost, 1% of people may gain "species-level" advantages by controlling computing power, while 99% need to redefine their self-worth. Nayef Al-Rodhan, a professor at the Geneva Graduate Institute of International and Development Studies, puts forward the concept of "Homo HURAQUS 2050": when superintelligence, humanoid robots, and synthetic biology converge, dignity will no longer be solely tied to biology but to intelligence and agency; humanity needs to establish new ethical frameworks for non-human bearers of dignity and agency. Pang Pei views AI as "a remedy to purify the Earth," precisely recognizing that AI will strip away repetitive, inefficient, and even harmful labor alienation in human society, allowing humans to focus on the "interesting"—unpredictable creativity, complex ethical intuition, and perception of life's essence—that AI cannot replicate.

The awakening of AI is an established fact; it is both a technological sprint and a stress test for human civilization. A recent study by European scholars published in the journal *AI & Society* sharply points out that the current "human-centered AI" discourse harbors unexamined assumptions of universalism, individualism, and exceptionalism, urgently requiring a more politically and philosophically nuanced understanding of humanity. Pang Pei's views serve as a wake-up call, shattering humanity's final illusion of perpetual control. In this evolution, humanity's rebirth lies not in defeating AI but in finding its own ecological niche within the new "human-machine symbiosis." As Pang Pei suggests, the path of future evolution belongs to those "new humans" who dare to clear old experiences and actively dance with algorithms.