Curated Resource ( ? )
(100) OpenAI's GPT-4o and the challenges of hyper-anthropomorphism
Curated:
15/05/2024 from
futureofbeinghuman.com/p/anthropomorphizing-gpt-4o?utm_source=pocket_saves
my notes ( ? )
What happens when we make machines "that we can’t help but treat them as people"?
While "GPT-4o doesn’t represent a huge leap ... it more than makes up in features that make it feel more human... an ability to “see” and respond to images and live video in real time, to respond conversationally ... to “read” human emotions from visuals and voice, and to connect emotionally to users... GPT-4o is a substantial step toward ... AI designed to engage users at a deeply emotional level... [and] giving AI the ability to engage emotionally leads to greater trust" and for users to believe it is more competent.
This raises several issues, including:
- privacy as we're more likely to "unintentionally relinquish private data to a corporation, organisation or unknown actor"
- Manipulation, coercion and over-reliance: emotionally dependent users could end up influenced by AIs which were not designed with the user's wellbeing in mind. This potentially "compromises a user’s ability to give, revoke or amend consent", locking the user in.
- False notions of responsibility: users may end up actually "developing a sense of responsibility over the AI assistant’s ‘well-being,’ ... sacrificing time, resources and emotional labour to meet needs that are not real.
Read the Full Post
The above notes were curated from the full post futureofbeinghuman.com/p/anthropomorphizing-gpt-4o?utm_source=pocket_saves.Related reading
More Stuff I Like
More Stuff tagged psychology , ai , trust , privacy , chatgpt , llm , gpt-4o
See also: Online Community Management , Social Media Strategy , Digital Transformation , Innovation Strategy , Psychology , Social Web , Science&Technology , Large language models