Is the ChatGPT interface you use every day about to change again?
Who writes the news content you read through AI search? Are the writers being paid?
What does it mean when AI companies say they're close to creating self-improving AI?
AI capabilities are expanding, but has there been clear discussion on "who owns the content and who is responsible if something goes wrong"?
OpenAI plans to combine ChatGPT's Chat and Work into a single interface.
Several news organizations, including The New York Times, are suing the Brave browser for unauthorized use of their reporting for AI purposes.
Anthropic states in its SEC filing that it is very close to achieving self-improving AI.
Today’s three major AI stories reflect issues around "product use," "content copyright," and "AI safety," highlighting how responsibility should be assigned as AI capabilities grow.
1. ChatGPT to Merge Chat and Work: What Will Change in the Interface?
What’s changing?
According to 9to5Mac, OpenAI will first invite a small group of users to test combining ChatGPT’s “Chat” and “Work” features into one interface.
After merging, you will be able to move seamlessly from chatting or brainstorming ideas to having ChatGPT execute tasks like building a website or creating a presentation—all without switching tabs.
However, the programming tool Codex will remain a separate desktop app.
What was Work originally?
Work refers to ChatGPT’s feature aimed at "task execution," capable of handling multi-step jobs rather than simple Q&A.
9to5Mac also noted that OpenAI President Greg Brockman previously expressed hopes to retire the standalone Work tab by the end of this year and merge its capabilities back into the regular ChatGPT.
When will the old app be retired?
The old desktop app has been renamed “ChatGPT Classic,” and a banner inside reads: This app will be discontinued on January 30, 2027.
The banner also says that your full chat history will transfer to the new app.
The new desktop app already integrates ChatGPT, Work, and Codex into one interface.
What do I need to do now?
If you’re using the old desktop ChatGPT app, be sure to download the new version from the official ChatGPT page before January 30, 2027.
If you primarily use the web or mobile app, current reports only mention the desktop app changes; there’s no news yet on adjustments for web or mobile versions.
The Chat and Work merger is still in small-scale testing and not visible to all users yet.
Is this related to the usage limit changes coming in December?
They are related but separate issues.
SasaDaily reported on November 2 that starting December, ChatGPT and Codex will share the same usage quota.
This update concerns the interface merger. Together, these changes show OpenAI’s intent to evolve ChatGPT from a "chat tool" into a platform where you can accomplish tasks through one unified entry point.
Sources:
2. New York Times and Others Sue Brave: Should AI Pay for Using News Content?
Who is suing whom?
On November 2, The New York Times, New York Daily News, Chicago Tribune, and Orlando Sentinel filed lawsuits against Brave Software in U.S. federal court in Manhattan.
Brave is a browser and search engine company that markets itself as privacy-focused.
What are the publishers accusing?
According to TheWrap, the main accusations are:
- Brave copied news content without authorization or payment and profited from it.
- Brave allows users to bypass publishers’ paywalls.
- Brave provides publisher content through its “Brave Search API,” which many AI companies use to scrape web content.
- Brave’s AI Q&A feature, Ask Brave, reportedly reproduces large excerpts of original articles, such as one Chicago Tribune article exceeding 1,200 words.
The New York Times also said it sent two registered letters to Brave’s CEO in August 2025 and April 2026 without response.
Why are AI companies involved?
An API is essentially a “data gateway” that allows other companies’ programs to directly access Brave’s search results.
The lawsuit names AI companies that have used the Brave Search API including Anthropic, Perplexity, Mistral, xAI, and Cohere.
Although only Brave is being sued, these AI companies are not defendants. However, the publishers’ lawyers say plainly: “Buying stolen goods doesn’t make them legal.”
What is Brave's response?
TheWrap reported that Brave had not responded to requests at the time of reporting.
Previously, Brave has argued that its search practices fall under “fair use” according to U.S. copyright law.
Brave is also facing similar lawsuits involving News Corp.
Does this mean Brave has broken the law?
No, it does not.
At this stage, these are allegations from the publishers and no court ruling has been made.
The core legal question about whether AI and search engines’ use of news content constitutes fair use is still highly contested and unresolved.
How does this affect everyday readers?
When you read summarized news directly from AI search or AI Q&A, you might not visit the original news sites.
If news sites lose subscription and advertising revenue because of this, they may reduce investigative reporting long-term.
This is why more publishers are suing AI companies or seeking licensing agreements.
Sources:
3. Anthropic’s SEC Filing Claims Nearing Self-Improving AI: Warning or Promotion?
What did Anthropic say?
Techzine quoted TechCrunch reporting that on November 2, Anthropic updated its SEC IPO filing, stating in the “Risk Factors” section that the company is very close to developing AI models capable of self-improvement.
The filing also clarifies that no model has yet crossed this threshold.
Techzine noted that the filing mentions using AI to develop stronger AI poses safety risks.
Note: SasaDaily has not accessed the original filings; this summary is based on Techzine’s reporting.
What is “self-improving AI”?
Also called Recursive Self-Improvement (RSI), it means AI can help train and improve the next generation of stronger AI, which then improves the subsequent generation, creating an accelerating cycle.
Supporters see it as a path to rapid technological breakthroughs; skeptics worry it could accelerate beyond human monitoring and control.
How does Anthropic plan to manage this?
Techzine says Anthropic will follow its “Responsible Scaling Policy” to increase monitoring and mitigate risks as model capabilities near this threshold.
The company also states humans will still be involved in the AI-driven research and development process in the near term.
Why include this in the IPO filing?
When preparing to go public, companies must disclose potential risks to investors.
So this is a “risk disclosure” rather than a product announcement.
Techzine also points out some skeptics see this as a marketing tactic.
Both interpretations exist, but it’s currently unclear which is more accurate.
Does this mean AI is already out of control?
No.
The filing says “close to” and clearly states no model has yet crossed the line.
You shouldn’t conclude just because a company says “close” that self-improving AI exists or is imminent.
The notable point is this risk is formally acknowledged by the AI company in a legal document.
Sources:
4. Why Should We Consider These Three News Stories Together?
One is about an app interface, another about a news lawsuit, and the third about AI safety. They seem very different.
But they all address the same question:
As AI does more, how do we divide responsibility?
- ChatGPT is evolving from a chat tool to one that does actual tasks, so someone must be responsible for vetting the outputs.
- AI uses news written by others—should the writers be compensated? Courts are adjudicating.
- AI is starting to develop AI—who ensures the process is safe? Even AI companies list this as a risk.
As capabilities advance, rules and responsibilities must keep pace. These stories serve as a reminder.
5. What Should the Average Person Watch for Today?
If you use ChatGPT on your computer:
Check if you’re using the old “ChatGPT Classic” app. If so, switch to the new version before January 30, 2027.
There’s no rush now, but don’t wait until the last minute.
If you often use AI search to read news:
AI-generated summaries are convenient, but for important matters, always click through to the original source for the full story.
Summaries may omit key context. The original source remains the responsible party.
If you see headlines saying “AI is about to go out of control”:
First clarify whether it’s “already happened,” “companies say they’re close,” or just “speculation.”
No need to panic or ignore; keep following trusted sources.
Most Important Point Today
The more AI does, the more we need to clarify: who owns the content, and who is responsible for the outcomes.
Recommended Reading
- Today’s AI Tools|2026/07/29: ChatGPT Work connects documents and work tools, turning research, organization, creation, and updates into a complete workflow
- AI Q&A|2026/10/09: Just because an article appears on a news website, does it mean the author is real and the content trustworthy?
- AI Q&A|2026/08/08: Does an AI agent stopping during testing mean it’s safe?