AI News: Everyone's Leaving ChatGPT!

Value of the Information & Strength of the Argument

The video provides a high-value, up-to-date summary of the week’s most important AI developments, with a clear focus on the OpenAI/Anthropic/Pentagon saga. The creator’s argumentation is generally solid, presenting facts from official sources and reputable media, while clearly labeling his own opinions and speculations. He effectively contextualizes the model releases, explaining their practical significance for different user groups. The coverage of the Pentagon drama is balanced, presenting multiple perspectives and highlighting the apparent contradictions in OpenAI’s actions. The video’s strength lies in its ability to synthesize complex information into an accessible narrative, though it occasionally relies on subjective impressions of model performance.

Scientific Rigor, Source Quality, Title Accuracy

The video demonstrates strong scientific rigor by consistently citing official sources (OpenAI, Google, Anthropic blogs) and reputable tech news outlets (TechCrunch, Bloomberg, BBC, Engadget). The creator provides links to all referenced articles in the description, allowing viewers to verify claims. The title accurately reflects the video’s main focus on the user exodus from ChatGPT, which is a central theme. The content is well-structured with clear chapters, and the creator distinguishes between factual reporting and his own analysis. The main weakness is the potential for bias in the selection and framing of news, but this is inherent to the news commentary format.

The title accurately reflects the video's main focus on the user exodus from ChatGPT to Claude following the OpenAI-Pentagon deal, while also covering other significant AI news.

8/10

The video is a well-structured weekly AI news roundup, presenting information from official sources (OpenAI, Google, Anthropic) and reputable tech media (TechCrunch, Bloomberg, BBC). The creator clearly separates facts from opinions and provides context for each announcement. The main limitation is the inherent bias of a news commentary format, where the creator's perspective shapes the narrative, and some claims (e.g., model performance) are based on demos and subjective impressions rather than independent benchmarks.

• 00:00Intro
• 00:14GPT-5.3 Instant
• 01:40GPT-5.4
• 09:40Box AI
• 10:56Gemini 3.1 Flash-Lite
• 12:37NotebookLM Cinematic Video Overviews
• 16:00Google Canvas in AI Mode
• 16:48Anthropic/OpenAI/Pentagon Drama
• 29:02Qwen 3.5
• 29:33Grok 4.20 Beta 2
• 29:50Microsoft Phi-4-reasoning-vision
• 30:25OpenAI Codex app in Windows
• 30:57Meta AI Glasses Exposed
• 33:20Be Inaudible Spectre I
• 34:45NVIDIA GTC 2026 + Giveaway

• GPT-5.3 Instant — Official OpenAI announcement of the GPT-5.3 Instant model.
• Introducing GPT-5.4 — Official OpenAI announcement of the GPT-5.4 model.
• Gemini 3.1 Flash-Lite — Google blog post introducing the Gemini 3.1 Flash-Lite model.
• NotebookLM Cinematic Video Overviews — Google blog post announcing the cinematic video overviews feature in NotebookLM.
• Canvas AI Mode — Google blog post about using Canvas in AI Mode for writing and coding.
• Anthropic Hegseth Statement — Anthropic's official statement regarding comments from the Secretary of War.
• OpenAI DoW Agreement — OpenAI's official announcement of its agreement with the Department of War.
• Claude Import Memory — Claude's feature page for importing memory from other AI providers.
• Claude Free Memory — X post from Claude AI announcing free memory availability.
• Ditching ChatGPT for Claude — TechCrunch article about users switching from ChatGPT to Claude.
• Claude Switching Thread — X thread discussing the switch from ChatGPT to Claude.
• ChatGPT Uninstalls Surged — TechCrunch article reporting a surge in ChatGPT uninstalls after the DoD deal.
• Anthropic $20B Revenue — Bloomberg article on Anthropic's revenue run rate amid the Pentagon feud.
• Anthropic Revenue Chart — X post showing an Anthropic revenue chart.
• Altman Defends Pentagon Work — WSJ article on Sam Altman defending OpenAI's Pentagon work to staff.
• Altman Pentagon No Say — Bloomberg article on Altman telling staff OpenAI has no say over Pentagon decisions.
• OpenAI Amends DoD Deal — Engadget article on OpenAI amending its DoD deal to prevent mass surveillance.
• Anthropic Trump Criticism — The Information article on Anthropic's CEO criticizing OpenAI's Pentagon deal as safety theater.
• FT AI Article — Financial Times article on AI and the Pentagon situation.
• Anthropic Supply-Chain Risk — Bloomberg article on the Pentagon designating Anthropic as a supply-chain risk.
• Qwen 3.5 Small Models — X post from Alibaba Qwen announcing Qwen 3.5 small models.
• Grok 4.20 Beta 2 — X post from Grok announcing Grok 4.20 Beta 2.
• Phi-4 Reasoning Vision — Microsoft Research blog post on Phi-4 Reasoning Vision.
• Codex Windows App — X post from OpenAI Developers announcing the Codex Windows app.
• Meta Glasses Privacy — Swedish newspaper article on Meta AI glasses and data privacy concerns.
• ICO Meta Glasses — BBC article on the ICO's investigation into Meta glasses.
• Meta Glasses Lawsuit — TechCrunch article on a lawsuit against Meta over smartglasses privacy concerns.
• Spectre I Anti-Recording — X post about the Spectre I anti-recording device.
• Daniel Eckler Profile — X profile of Daniel Eckler, likely the creator of Spectre I.

• TechCrunch: Users are ditching ChatGPT for Claude — Corroborates the video's claim of a user exodus from ChatGPT to Claude.
• TechCrunch: ChatGPT uninstalls surged by 295% after DoD deal — Provides data on the surge in ChatGPT uninstalls, supporting the video's narrative.
• Bloomberg: Anthropic nears $20 billion revenue run rate — Confirms Anthropic's strong financial position despite the Pentagon conflict.
• Engadget: OpenAI will amend defense department deal to prevent mass surveillance — Supports the video's claim that OpenAI amended its DoD deal to address surveillance concerns.

• The Information: Anthropic CEO told employees OpenAI Pentagon deal is 'safety theater' — This source suggests that Anthropic's CEO views OpenAI's Pentagon deal as a superficial safety measure, contradicting the video's more neutral presentation of OpenAI's red lines.

• futuretools.io
• futuretools.io
• www.linkedin.com
• www.threads.net