GPT-5 Has the Brains, But Where’s the Heart? Altman Promises a Softer Touch

OpenAI CEO Sam Altman has responded to a wave of user backlash following the rollout of GPT-5, pledging to make the new model more emotionally engaging and customizable. While GPT-5 boasts significant technical improvements, many users have lamented the loss of charm, wit, and relatability that made earlier versions like GPT-4o so beloved.

The GPT-5 Backlash: What Sparked the Outcry

Soon after GPT-5 launched, OpenAI discontinued access to older models including GPT-4o, GPT-4.1, and GPT-4.5. This abrupt shift left users with no option to interact with their preferred versions, triggering widespread frustration.

Key highlights:
- GPT-5 was promoted as faster, more accurate, and less prone to hallucinations or excessive agreement
- Users reported shorter, colder, and less emotionally nuanced responses compared to GPT-4o
- Social media erupted with complaints about the model’s lack of personality and warmth

Altman acknowledged that his team underestimated how much users valued the emotional intelligence and creativity of GPT-4o. In a public statement, he admitted that while GPT-5 performs better in many technical areas, it lacks the human-like qualities users had grown attached to.

OpenAI’s Immediate Response and Long-Term Vision

To address the concerns, OpenAI has taken several steps and outlined a roadmap for future improvements.

1. Restoring GPT-4o access for ChatGPT Plus subscribers, though free users remain limited to GPT-5
2. Committing to finish GPT-5’s rollout and stabilize its performance before implementing personality upgrades
3. Promising deeper customization options so users can tailor the AI’s tone, style, and emotional depth

Altman emphasized that users have vastly different preferences. Some want humor and emojis, others prefer cold logic and professionalism. OpenAI’s long-term goal is to offer steerable personalities that adapt to individual user needs.

Customization and Personality: The Next Frontier

OpenAI is investing in research to allow users to personalize their AI experience more deeply. This includes:
- Personality previews that let users choose between warmth, logic, creativity, or other traits
- Interface updates to make it easier to identify which model is responding
- Fixes to technical glitches, including a broken autoswitcher that made GPT-5 seem less intelligent

Altman noted that truly effective customization will take time, but the company is committed to transparency and rapid iteration. He also acknowledged upcoming capacity challenges, promising to explain any trade-offs made to maintain performance.

Why This Matters: The Human Touch in AI

The GPT-5 controversy highlights a growing tension in AI development: technical excellence versus emotional resonance. While GPT-5 excels in coding, reasoning, and multi-modal tasks, users still crave the subtle human-like qualities that make conversations feel authentic.

Key takeaways:
- GPT-5’s rollout marks a turning point in how users evaluate AI—not just by accuracy, but by personality
- OpenAI’s response reflects a broader industry shift toward emotionally intelligent machines
- Altman’s promise to make GPT-5 warmer could redefine what users expect from conversational AI

As OpenAI works to balance performance with charm, the future of GPT-5 may hinge not just on how smart it is—but on how well it listens, laughs, and connects.

Sources: Mint, CNBC TV18, Financial Express, The Hans India