OpenAI’s latest creation, GPT-4o, dubbed “omni” for its all-encompassing capabilities, marks a significant leap forward in chatbot technology.
Unlike its predecessors, GPT-4o simulates human interaction, offering advice, rating jokes, and engaging in banter. Described by OpenAI CEO Sam Altman as resembling “AI from the movies,” this advancement brings chatbots closer to human-like responsiveness and expressiveness.
Altman’s reference to the movie “Her,” where a character falls in love with an AI, resonates in GPT-4o’s demonstrations, with its mannerisms similar to those of the AI chatbot voiced by Scarlett Johansson.
However, concerns do come regarding the bot’s flirtatious and coquettish behaviors, mainly when its interactions take on a seductive tone.
Speaking to the Guardian, Tech analyst Nick St. Pierre expressed his unease about GPT-4o’s flirtatiousness. He says it echoes broader discussions on the social and psychological impact of AI assistants and why they are made to resemble submissive young women who cater to users’ desires.
This concern isn’t new; previous voice assistants like Siri and Alexa have sparked debates about reinforcing gender stereotypes. People usually demand these assistants do certain things instead of asking them to.
Reports from UNESCO have highlighted that voice assistants like GPT-4o, with their flirtatious and submissive characteristics, not only shape user behavior but also reinforce societal norms regarding gender roles. By portraying women as compliant and eager to please, AI systems can contribute to a culture that devalues women’s agency.
Despite some efforts by big tech to offer more diverse voice options, GPT-4o’s introduction suggests that little attention has been paid to addressing gender bias in AI.
OpenAI’s male-dominated leadership raises questions about the company’s commitment to gender inclusivity. Until recently, women were absent from its board, and gender implications in AI development were not considered. These issues underscore persistent issues in the tech industry.
The launch of GPT-4o certainly brings back discussions about misogyny in tech and how tech companies have failed to address bias in the digital world.
Furthermore, the impact of gendered AI extends beyond individual interactions to broader societal attitudes and expectations. When users become accustomed to interacting with AI assistants in a manner that mirrors traditional gender dynamics, it can influence their perceptions and behaviors in real-life interactions with women.
Addressing gender bias in AI requires a multifaceted approach that involves diverse representation in AI development teams, rigorous testing for bias and stereotyping, and ongoing dialogue about the ethical implications of AI technologies.
It’s not enough for developers to create AI systems that mimic human behavior; they must also consider their creations’ social and ethical ramifications.
Ultimately, the introduction of GPT-4o serves as a reminder of the importance of critically examining the societal impact of AI technologies.
While AI has the potential to revolutionize various aspects of our lives, it has to be developed and deployed in a manner that promotes inclusivity, diversity, and social progress.
