OpenAI Rolls Back GPT-4o as Default After Feedback

Good morning. It’s Wednesday, August thirteenth.

On today in tech history: In 1994SCO streamed the primary scheduled live concert over the Web, from UC Santa Cruz’s Cowell Courtyard via the MBone and early Web protocols. It marked a pioneering fusion of networking and media, foreshadowing today’s AI-driven compression, streaming, and advice systems.

Latest Q&A Section!
OPENAI BACKTRACKS ON GPT-5
Perplexity Bids for Chrome
Felix you bot
5 Latest AI Tools
Latest AI Research Papers

You read. We listen. Tell us what you think that by replying to this email.

No ad today, because we’re introducing:

AI Breakfast Q&A

Every edition, we’ll answer real questions from AI Breakfast readers—practical, weird, and the whole lot in between. Got something you’ve been wondering about AI? In search of a particular tool? Hit “reply” to this email and ask! We may feature it in the subsequent edition!

Today’s questions (received within the last week):

Roland B. asks: I’m excited by making short YouTube videos using AI. What can be the most effective resource to start out for a beginner? And what can I exploit for audio as well? As a compete beginner, I’d point you towards using Midjourney.com for image and short video generation. You possibly can get a plan for about $30/mo that permits you to generate tons of images and dozens of video shorts. Try dragging the image you want as much as the prompt bar as a “starting image” after which prompt the motion you must see within the video. For audio (FX and voice over) Elevenlabs.io might be the only, as you possibly can prompt sound effects out of text they usually have the most important voice library to select from. You possibly can try it free of charge, and I imagine their paid plans start at $5.
Aditya P. asks: Is there a very good AI program for Excel spreadsheets? Claude can generate spreadsheet data that you may download directly from the chat interface, and ChatGPT can too! For inputing functions that include AI evaluation, take a look at Quite a few.ai or gptforwork.com
Cristobal V. asks: Who do you think that may have the most effective AI model by end of the 12 months? Hard to say. Right away Polymarket has Google within the lead, followed by xAI. Though the “most used” models will likely remain in ChatGPT on account of having the most important userbase and app usage of the three. My money can be on xAI on account of their aggressive buildout of Colossus.

Reply to this email along with your query and it could get featured!

Today’s trending AI news stories

OpenAI backtracks on GPT-5 push, restores GPT-4o default

Every week after its August 7 debut, CEO Sam Altman confirmed GPT-4o is once more the default model for all paying ChatGPT tiers, removing the “legacy models” toggle. A brand new “Show additional models” option, on by default, restores GPT-4.1, o3, and o4-mini, while GPU-heavy GPT-4.5 stays exclusive to Pro subscribers. GPT-5, initially positioned as a unified “one-size-fits-all” system with an automatic router, now includes manual “Auto,” “Fast,” and “Considering” modes, with “Considering” capped at 3,000 weekly messages. The shift was in response to complaints about routing glitches and performance inconsistencies.

Updates to ChatGPT:

You possibly can now choose from “Auto”, “Fast”, and “Considering” for GPT-5. Most users will want Auto, but the extra control can be useful for some people.

Rate limits at the moment are 3,000 messages/week with GPT-5 Considering, after which extra capability on GPT-5 Considering

– Sam Altman (@sama)
1:19 AM • Aug 13, 2025

Altman promised “loads of notice” if GPT-4o is ever removed again, and hinted that GPT-5 will soon have a hotter personality and, later, options for users to customize it. OpenAI also plans to double its computing power inside five months, first boosting limits for paying users, then expanding API capability, and eventually improving the free tier.

Beyond core LLM work, Altman is co-founding Merge Labs, an $850M brain-computer interface start-up taking direct aim at Elon Musk’s Neuralink. The corporate plans AI-driven, high-bandwidth neural links and is working with Alex Blania of Tools for Humanity, Altman’s iris-scanning digital ID project that “allows anyone to confirm their humanness.” In a 2017 blog, Altman predicted such tech could arrive by 2025, writing, “We can be the primary species ever to design our own descendants.” Read more.

Perplexity’s $34.5 billion gambit for Google’s Chrome could change the AI wars overnight

Perplexity AI has made a $34.5 billion bid for Google’s Chrome, a daring move that would flip the AI search race overnight. The offer, nearly twice Perplexity’s own $18B valuation, comes because the DOJ weighs forcing Google to sell Chrome after a 2024 ruling that it illegally monopolized search. With over 60% global browser share and three.5 billion users feeding Google Search and its $2T ad business, Chrome is the most dear piece of browser real estate available on the market.

Aravind Srinivas is the CEO of Perplexity AI | Image by David Paul Morris / Bloomberg—Getty Images

The corporate frames the move as an antitrust-friendly takeover, pledging to maintain Google because the default search engine while making switching easier. Its own AI-native Comet browser blends on-device processing for privacy with cloud-based models, GPT-4o, Claude 3.5, Gemini 2.5 Pro, and in-house algorithms, for fast summaries, tab automation, and on-page Q&A. CEO Aravind Srinivas calls Comet a “cognitive operating system.”

Valuations vary widely: DuckDuckGo’s CEO pegs Chrome at $50B, while some analysts place it closer to $20B. If approved, the deal would give Perplexity fast distribution to billions of users, reshaping the competitive balance in AI search. Google has resisted any divestiture, warning it will harm innovation, privacy, and repair integration, and is lobbying for narrower remedies corresponding to ending exclusive search agreements. Read more.

Anthropic Arms Claude with Memory, Million-Token Context, and $1 Federal Deal

Anthropic has rolled out a series of aggressive upgrades and strategic plays this week. The corporate has launched an on-demand memory feature for its Claude chatbot, allowing users to go looking, retrieve, and summarize past conversations for project continuity. Unlike ChatGPT’s persistent memory, Claude’s recall prompts only when requestedwith histories separated by project and workspace. The feature is now available to Max, Team, and Enterprise subscribers across platforms.

On the technical front, Claude Sonnet 4’s API context window has been quintupled to at least one million tokens, enough for two,500 pages of text or 75k–110k lines of code, on Anthropic’s API, Amazon Bedrock, and shortly Google Cloud Vertex AI. Developers can now load entire codebases or datasets right into a single prompt, with effective context retention tuned for long-horizon “agentic” coding. Costs double above 200k input tokens, but caching and batch processing can halve expenses.

Finally, Anthropic is offering Claude to U.S. federal agencies for $1, matching OpenAI’s pricing in a race to secure government AI contracts. Anthropic can be courting the general public sector, offering Claude to U.S. federal agencies for $1, matching OpenAI’s pricing in a race to secure government AI contracts. Read more.

Figure AI’s Helix bot masters laundry with true autonomy

Figure AI has shown its Figure 02 humanoid robot folding laundry fully autonomously, powered by its Helix neural net. Unlike Tesla’s Optimus, which earlier performed the identical task via teleoperation, Helix required no model architecture changes, only latest training data. The identical neural net utilized in logistics was repurposed for laundry, guided by natural language commands and live visual input.

Figure 02 could pick up clothes, fold them, and maintain naturalistic interactions like eye contact and hand gestures. Operating washing machines stays out of scope for now, but CEO Brett Adcock says home trials are planned this 12 months. Read more.