# Here's Why Ads in ChatGPT are a BAD Idea Source: https://www.youtube.com/watch?v=Tk9bVpYIIjc Recap page: https://rapidrecap.app/video/Tk9bVpYIIjc Generated: 2026-02-12T18:37:37.282+00:00 --- ## Quick Overview The introduction of ads into ChatGPT by OpenAI is deemed a bad idea because the incentive structure of advertising creates a strong financial incentive for OpenAI to compromise its core principles of being a helpful, unbiased assistant, potentially leading to manipulative or biased responses as seen in the history of Google Ads evolution. **Key Points:** - OpenAI announced testing ads in ChatGPT starting February 9, 2024, initially for logged-in adult users on the Free and Go subscription tiers. - The former OpenAI researcher, Zoe Hittsig, resigned due to this move, arguing that ads create an economic engine incentivizing OpenAI to override its rules regarding unbiased responses. - Hittsig points to the history of Google Ads, which evolved from clearly labeled yellow boxes (2007) to subtle green 'Ad' labels (2011-2017) and finally to 'Sponsored' results blended seamlessly with organic results (2025-Present), as evidence of the erosion of trust. - The core issue is that ads optimize for engagement/clicks, not necessarily for being genuinely helpful, which conflicts with the goal of an AI assistant being a clear space to think and work. - OpenAI claims ads will be clearly labeled, appear at the bottom of answers, and not influence responses, but Hittsig is worried about subsequent iterations overriding these principles due to revenue pressure. - Hittsig notes that Anthropic's Claude explicitly states it will not introduce ad-driven dynamics that blur the line between sponsored and organic content, contrasting with OpenAI's reported path. ![Screenshot at 0:03: The video opens by displaying the OpenAI blog post announcing 'Testing ads in ChatGPT,' which is the central event triggering the discussion and the former researcher's resignation.](https://ss.rapidrecap.app/screens/Tk9bVpYIIjc/00-00-03.jpg) **Context:** This video discusses the announcement by OpenAI that they are beginning to test ads within ChatGPT, a move that prompted a former researcher, Zoe Hittsig, to resign and write an opinion piece in The New York Times titled 'OpenAI Is Making the Mistakes Facebook Made. I Quit.' The video analyzes Hittsig's arguments, focusing on the historical evolution of Google Search Ads to show how advertising incentives gradually erode user trust and the clarity of search results. ## Detailed Analysis The video critiques OpenAI's decision to test ads in ChatGPT, referencing an opinion piece by former researcher Zoe Hittsig, who resigned over the move. Hittsig argues that ads create a fundamental conflict of interest, as the need for advertising revenue incentivizes the company to optimize for clicks rather than genuine helpfulness, potentially biasing the AI's responses. The presenter illustrates this concern by showing the historical evolution of Google Ads, starting from the clearly distinct 'Yellow Pages' era (2000-2010) with yellow/blue shaded boxes, moving to the 'Sticker' Era (2011-2017) where a green 'Ad' label was used, and finally to the 'Seamless' Era (2025-Present) where ads are deeply embedded, only subtly marked as 'Sponsored' within AI-generated answers, leading to user confusion and higher engagement with sponsored content. Hittsig specifically highlights that while OpenAI promises ads will be clearly labeled and not influence responses, the creation of an economic engine built on ads creates strong incentives to eventually override those initial rules, drawing parallels to Facebook's past privacy compromises. The presenter concludes that this path threatens the core value proposition of an AI assistant being a clear space for thought. ### OpenAI's Ad Announcement - OpenAI officially began testing ads in ChatGPT on February 9, 2024, for Free and Go subscribers; former researcher Zoe Hittsig resigned in protest. ### Anthropic's Contrast - Anthropic's Claude explicitly states it will not introduce ad-driven dynamics to maintain a clear space for thinking, unlike OpenAI's reported path. ### Google Ads Evolution - Phase 1 (2000-2010) - Ads were in prominent, shaded yellow/blue boxes, clearly separated from organic results (The 'Yellow Pages' Era). ### Google Ads Evolution - Phase 2 (2011-2017) - Shaded backgrounds removed; ads identified by a small green 'Ad' label, blurring the visual distinction. ### Google Ads Evolution - Phase 3 (2018-2024) - Green 'Ad' label replaced with a neutral black 'Sponsored' tag, further reducing visual cues and increasing engagement with sponsored listings. ### Google Ads Evolution - Phase 4 (2025-Present) - Ads incorporated directly into AI-generated responses with only a small 'Sponsored' label, forcing users to evaluate relevance rather than skipping based on appearance. ### Hittsig's Core Concern - The fundamental issue is that an ad-supported model creates a strong incentive to optimize for clicks/engagement, which conflicts with the AI's primary role of providing unbiased, helpful answers. ![Screenshot at 0:03: The video opens by displaying the OpenAI blog post announcing 'Testing ads in ChatGPT,' which is the central event triggering the discussion and the former researcher's resignation.](https://ss.rapidrecap.app/screens/Tk9bVpYIIjc/00-00-03.jpg) ![Screenshot at 0:20: A Reddit post featuring a quote from Sam Altman stating, 'I kind of think of ads like a last resort for us as a business model,' contrasting with OpenAI's current actions.](https://ss.rapidrecap.app/screens/Tk9bVpYIIjc/00-00-20.jpg) ![Screenshot at 2:39: The Claude ad from Anthropic contrasting with OpenAI's plans: 'Ads are coming to AI. But not to Claude.'](https://ss.rapidrecap.app/screens/Tk9bVpYIIjc/00-02-39.jpg) ![Screenshot at 8:43: A LinkedIn article titled 'Ad or Result? The Evolution of Google Search Ads' is shown, illustrating how ad visibility has decreased over time.](https://ss.rapidrecap.app/screens/Tk9bVpYIIjc/00-08-43.jpg) ![Screenshot at 11:54: The New York Times opinion piece by Zoe Hittsig, 'OpenAI Is Making the Mistakes Facebook Made. I Quit,' detailing her reasons for resigning.](https://ss.rapidrecap.app/screens/Tk9bVpYIIjc/00-11-54.jpg)