EN

Sam Altman Says GPT-5 Is Already Here... (Sam x Theo Von)

#AI Tools
2025-07-29 20:5410 min read

Content Introduction

In this video, the speaker shares insights about their experience testing the new AI model, GPT-5. They reflect on the efficiency and capabilities of the AI, noting moments of surprise and a sense of inadequacy against such advanced technology. The speaker discusses the evolving landscape of AI, including removal of model selectors and the possibility of a universal model that integrates the best from various iterations. They express concerns over the economic implications of AI's growth, particularly in terms of power and income distribution, advocating for a model that ensures benefits extend to all. The conversation highlights the potential of AI to revolutionize tasks but warns against sacrificing human agency in the process. The nuances of privacy and increased surveillance are also considered, alongside reflections on historical benchmarks in technology and the importance of harnessing AI's potential judiciously.

Key Information

  • The speaker tested a new AI model called GPT5 and noted its ability to answer questions accurately.
  • Sam Alman, a key figure at OpenAI, is actively testing and using GPT5.
  • GPT5 is undergoing changes, including the removal of the model selector to utilize the best features from various models.
  • There's a rumored token limit of 1 million for the new model, which may enhance user experience.
  • The discussion also included the concept of universal basic income as a potential solution for economic disparities in an AI-driven future.
  • The speaker expressed concern over increased surveillance and privacy issues as AI becomes more pervasive.
  • The future AI landscape may require new economic models to ensure equitable distribution of wealth generated by AI technologies.
  • There is ongoing debate about whether AI should play a role in government systems and its implications for societal structure.

Timeline Analysis

Content Keywords

AI Testing

The speaker discusses testing a new AI model, GPT-5, and shares how it answered a complex question effortlessly, prompting reflections on the capabilities of AI versus human intelligence.

GPT-5 Features

Highlights include that GPT-5 is testable, with active testing by Sam Alman, and discusses improvements such as removing the model selector for better performance by choosing the best model based on input.

Token Limit

There is a rumored token limit of 1 million tokens for GPT-5, considered significant as users experienced this with previous releases like Gemini 2.5.

AI and Economic Models

The conversation revolves around AI's impact on the economy, discussing concepts like universal basic income and the need for new economic frameworks that allow citizens to benefit from AI advancements.

Surveillance Concerns

The speaker expresses concerns about increased surveillance as AI technologies advance, questioning the balance between security and privacy rights in society.

Future of AI

Discussion touches upon the potential future where AI takes a central role in economic decisions, influencing government policies and individual rights.

AI Personalization

The concept of GPT personalization is explored, including how the AI can learn from user interactions to enhance the user experience.

AI's Role in Innovation

The speaker speculates on AI's role in generating new scientific discoveries and innovations, emphasizing the necessity of human oversight in cultural and technological advancements.

Related questions&answers

What is GPT5?

GPT5 is a new model from OpenAI that is currently in testing and is described to answer questions effectively. It aims to combine the best features of previous models.

How does GPT5 differ from previous models?

GPT5 is designed to remove the model selector, potentially creating a single model that combines the best elements of all prior versions without requiring users to select from multiple models.

What are the expected features of GPT5?

GPT5 is rumored to support a token limit of 1 million tokens and possibly include video modality, similar to Gemini 2.5 Pro, allowing users to input multimedia content directly.

What are the privacy concerns with AI development?

As AI becomes more prevalent, there are concerns that it will lead to increased surveillance and a loss of privacy, necessitating a balance between technological advancement and individual rights.

What idea was presented regarding universal basic income?

The discussion suggested that Universal Basic Income (UBI) might become a necessity in an increasingly automated economy to ensure individuals have a financial safety net.

Will AI replace human jobs?

While AI can handle many tasks efficiently, there is a consensus that humans still need to have agency and play a role in decision-making processes, especially regarding the future and ethical considerations of AI use.

What role does Sam Alman play in AI development?

Sam Alman is a notable figure at OpenAI, actively testing GPT5 and contributing to discussions about future AI frameworks and ethical considerations.

What potential issues might arise from incorporating AI into society?

There are concerns about economic disparities, where those who own AI technologies may become disproportionately powerful compared to those who don't, which could lead to societal unrest.

What is the timeline for the release of GPT5?

The release of GPT5 is anticipated sometime between July and September, with the possibility of delays.

More video recommendations