Sam Altman Says It'll Take Another Year Before ChatGPT Can Start a Timer
openai
| Source: Mastodon | Original article
OpenAI’s chief executive Sam Altman told reporters that ChatGPT will not be able to start a reliable timer for another twelve months. The comment came during a live interview about the model’s voice capabilities, when Altman was asked why the assistant sometimes fabricates a “timer” response that looks plausible but never actually counts down. He said the underlying voice model simply lacks the real‑time control needed for such a function and that “maybe another year before something like that works well.” The admission sparked a wave of criticism on social media, with users highlighting the gap between the product’s marketing promises and its actual performance.
The revelation matters because it underscores a growing trust issue around large language models that are increasingly positioned as everyday assistants. When a $852 billion‑valued company like OpenAI advertises features that it cannot deliver, users may begin to doubt other, more consequential claims—ranging from factual accuracy to safety safeguards. The episode also revives concerns raised in our April 8 coverage of Altman’s influence over AI’s future, where we questioned whether the company’s rapid rollout pace compromises transparency and reliability.
What to watch next: OpenAI’s product roadmap is expected to detail a timeline for integrating real‑time functions into its next‑generation model, rumored to be called GPT‑5. Analysts will be looking for concrete milestones at the company’s upcoming developer conference, while regulators in the EU and the U.S. may intensify scrutiny of “hallucination” mitigation claims. Meanwhile, competitors such as Anthropic and Google DeepMind are likely to highlight their own timing or scheduling capabilities as a differentiator, potentially reshaping the competitive landscape for voice‑enabled AI assistants.
Sources
Back to AIPULSEN