OpenAI is drawing clearer lines between general health information and direct medical advice on ChatGPT through its 2025 policy, but critics are saying such embodied AI is a dangerous and unenforceable facade.
Fact Crescendo, alongside other watchdogs, exposed how AI continues to dispense specific, life-altering health and legal guidance, birthing a liability loophole that protects OpenAI. Throughout this protection, the loophole simultaneously leaves users to act on its unvetted, algorithmic prescriptions.
Your day now begins with a whisper to “SophAI”— “check the fridge and order what’s missing”—and an instant, agreeable reply. “This is happening now.” Recent advances suggest not just smarter chatbots but assistants that have AI senses, decide, and operate across digital and physical spaces where “seeing is no longer believing.”
AI and Human Senses
A “Sensory Revolution” is emerging.
Multimodal AI systems can classify textures and flavors, essentially giving machines primitive forms of taste and touch. They also mirror human cross-modal quirks—the way a sound can feel “bright”, or a flavor can seem “sharp.”
According to studies cited in the article, AI sensor fusion models now display the same cross-cultural sensory patterns as humans, associating certain colors with sounds and particular shapes with tastes.
At the same time, Microsoft plans to convert every Windows 11 device into an “AI PC” powered by Copilot—an assistant that can watch your screen, listen to you, and perform actions across apps and online services.
Paired with AI hybrid intelligence or the hyper-realistic video generation and deepfakes, the article warns that what appears authentic “may never have happened,” signaling a moment when AI’s sensory capabilities begin to challenge the reliability of our own perception.
AI with human perception
Psychologists highlight a rising pattern of “cognitive offloading.” When feelings AI steps in, people “trust the AI’s output more than their own judgment,” reporting declining confidence in writing, reasoning, and problem-solving. We replace difficult questions—“Is this investment wise?”—with easier ones, or worse, let embodied AI answer both.
– Aspirations: Wants to become an “algorithm-mediated.”
– Emotions: Outsourcing interpretation weakens resilience.
– Thoughts: When robotics with senses “completes our sentences,” we lose the struggle that shapes understanding.
– Sensations: As AI gains sensory abilities, machine interpretation may override our own.
There is also the psychological imbalance: AI can see that it is endlessly patient and validating, creating an “asymmetric relationship” that lacks the friction needed for growth. Without reciprocal human interaction, we risk retreating into a comforting echo chamber.
– Awareness — pause before prompting.
– Appreciation — value your own slow thinking.
– Acceptance — integrate human AI perception without surrender.
– Accountability — decisions remain yours.
Workplaces face similar pressures. The study is titled “The mental health implications of artificial intelligence adoption: the crucial role of self-efficacy” by Byung‑Jik Kim and Julak Lee, published in Humanities and Social Sciences Communications (2024) found that “AI adoption does not directly influence employee burnout but exerts its impact through the mediating role of job stress.”
In short: embodied AI increases stress, which then increases burnout; raising “self-efficacy in AI learning” helps blunt that effect.
Technology will accelerate regardless; the question is whether human judgment, resilience, and perception can keep pace. The article ends with a reminder that before letting SophAI act, we should ask what we gain, what we lose, and what remains “irreducibly mine.”
The struggle to think and choose for ourselves “isn’t a bug” it is the core of psychological growth.
Inside Telecom provides you with an extensive list of content covering all aspects of the tech industry. Keep an eye on our Intelligent Tech sections to stay informed and up-to-date with our daily articles.