- Horizon AI
- Posts
- OpenAI Introduces ChatGPT Pulse for Proactive Daily Briefings ⭐
OpenAI Introduces ChatGPT Pulse for Proactive Daily Briefings ⭐
How to delete your Google Gemini activity
Gina Acosta
September 26, 2025
In partnership with
Welcome to another edition of Horizon AI,
OpenAI is launching a new feature in ChatGPT that will generate personalized reports for you while you sleep.
Let’s check it out!
Read Time: 4.5 min
Here's what's new today in the Horizon AI
OpenAI Launches ChatGPT Pulse
Nearly All Devs Use AI—but Few Actually Trust It, Google Finds
AI Tutorial: How to delete your Google Gemini activity
AI Tools to check out
AI Findings/Resources
The Latest in AI and Tech 💡
AI News
OPENAI
OpenAI Launches ChatGPT Pulse
This new feature will have ChatGPT perform personalized research on your behalf overnight, delivering updates and reminders each morning based on your chats, feedback, and connected apps.
Details:
Every morning, you get personalized morning briefings in the form of “topical visual cards you can scan quickly or open for more detail, so each day starts with a new, focused set of updates.”
If you have Gmail or Google Calendar connected, Pulse may even draft a sample meeting agenda, remind you to buy gifts, or provide suggestions for an upcoming trip.
The feature is currently in preview and only available to Pro users.
OpenAI frames this as a step toward true AI agents that understand your goals and act without waiting for prompts: less typing, more timely help. CEO Sam Altman even called it his “favorite feature of ChatGPT so far.”
TOGETHER WITH JULIUS AI
Your AI Analyst in Seconds → Data Analysis Made Simple and Fast
Data analysis is supposed to help you make better decisions. But most tools make data analysis slow, complex, and frustrating while relying too heavily on specialists.
Now, there’s an AI that changes this.
Meet Julius — your AI data analyst.
Upload a file or paste a link. Ask in plain English. Julius cleans the mess, finds patterns, and returns charts, summaries, and next steps in minutes.
What you get:
Hours saved versus cleaning spreadsheets.
Clear visuals ready for decks.
One-click reports to share by link.
Threads for quick questions; Notebooks for repeat jobs.
Works on CSVs, Excel, and more—no code.
2M+ people already use Julius to unlock insights instantly.
Try it for free 👉 julius.ai
Nearly All Devs Use AI—but Few Actually Trust It, Google Finds
Google Cloud’s 2025 DORA report says AI is now part of daily life for developers, but belief in its answers lags far behind its use.
Details:
90% of developers use AI day to day, yet only 24% say they trust its output; many treat it like a brilliant but unreliable coworker.
Devs spend a median two hours with AI assistants and report big gains: 80% say efficiency improved and 59% see better code quality, even as 30% trust AI “a little” or “not at all.”
Reliance without trust is the paradox: 65% say they depend on AI tools, despite not fully trusting them. This echoes Stack Overflow’s 2025 survey, which found that distrust in AI accuracy rose from 31% to 46% in just one year, even as adoption climbed to 84%
Google goes beyond just documenting the trend and has introduced its DORA AI Capabilities Model, a framework that identifies seven practices designed to help organizations harness the value of AI without incurring risks.
Despite trust issues, the report suggests that developers have successfully leveraged AI to boost productivity. Google’s own metrics support this, with CEO Sundar Pichai noting a 10% productivity increase across engineering teams and reporting that more than a quarter of the company’s new code now comes from AI systems.
AI Tutorial
How to delete your Google Gemini activity
Sometimes you might make mistakes and leave personal information in your Gemini chats. In those cases, it’s useful to delete your recent activity from the app or website.
On the Website
Go to the Google Gemini website and sign in.
Click Settings and Help in the lower-left corner, and select the Activity (clock with arrow) icon.
This will take you to the Gemini Apps Activity page.
To turn off and delete all activity: choose Turn off and delete activity → click Next → confirm Delete.
To remove specific items: scroll down Gemini Apps Activity → click X next to an entry → confirm Delete.
To delete by time range: open the Delete dropdown → pick Last hour, Last day, All time, or Custom range → click Delete.
On the App
Open the Gemini app (Android or iOS).
Tap your Profile icon in the top right.
Select Gemini Apps Activity.
From here, the steps are the same as on the website (turn off, delete specific entries, or set a time range).
*Note: By default, Google stores your Gemini Apps activity in your Google Account for up to 18 months. However, this only applies if you’re 18 or older; for users under 18, it’s off by default.
AI Tools to check out
📽 VidHex: All-in-one video solutions powered by AI.
✉️ Envelope: First AI agent for event planning. Describe your event and get a stunning registration page, invites & more.
📹 Lucy Edit AI: Edit videos using simple text prompts.
👀 Imini: A super AI agent that supports LLM conversations and can generate slides, AI-powered documents, and images/videos with a single prompt.
💥 Vuepak: Supercharge your outreach with AI, multichannel automation, and engaging multimedia.
AI Findings/Resources
😂 A man asks ChatGPT to guess the weight of random objects... and punishes it when it gets them wrong
💡 AI Crash Course to help busy builders catch up to the public frontier of AI research in 2 weeks
👉 4 ways to use Perplexity that have nothing to do with research
👀 How developers are using Apple’s local AI models with iOS 26
The latest in AI and Tech
The AI startup claims that OpenAI stole trade secrets by hiring former xAI employees who had knowledge of the company’s Grok chatbot and business plans.
OpenAI denies the claims, calling the lawsuit part of Musk’s “ongoing harassment.”
The UK government says a new AI system helped recover £480m in the past year, the largest anti-fraud haul on record.
More than £180m came from Covid loan scams, with other funds reclaimed from council tax fraud and illegal subletting.
The tool, called the Fraud Risk Assessment Accelerator, will now be licensed abroad, with the US, Canada, Australia, and New Zealand expected to adopt it.
The platform is rolling out new policies to fight AI-generated spam, impersonation, and undisclosed use of AI in music.
The company is working with DDEX to set a metadata standard requiring AI disclosure, backed by 15 labels and distributors. It’s also banning unauthorized voice clones and launching a spam filter to block tactics like short-track uploads.
Meta is putting “tens of millions” into a new super PAC, the American Technology Excellence Project, to oppose state-level AI regulations.
According to the company, the PAC will advocate for AI progress, defend U.S. tech leadership, and emphasize parental control amid rising child safety concerns.
Indeed’s latest AI at Work Report finds 26% of jobs posted in the past year could be “highly” transformed by generative AI, though only 0.7% of job skills are at high risk of full replacement.
The report suggests AI adoption will vary by industry and company, with transformation, not replacement, being the dominant trend.
That’s a wrap!
We'd love to hear your thoughts on today's email!Your feedback helps us improve our content |
Not subscribed yet? Sign up here and send it to a colleague or friend!
See you in our next edition!
Gina 👩🏻💻