🤖 ChatGPT Prompt Of The Day: Simplify complex sentences
✍️ Write Better Prompts with Copilot!
Microsoft’s latest features will make you a prompt engineering pro in no time!
Soon, Copilot for Microsoft 365 will sport an auto-complete feature that offers suggestions to improve your AI prompts.
For example, if you start creating a prompt and type “summarize,” Copilot will display options to summarize the last 10 unread emails in your inbox or other tasks related to your Office data.
Copilot will also 'elaborate' your prompts, making them richer with the click of a button.
This will definitely come in handy for those times when you want Copilot to take action on a file but are not sure what level of detail will help you get the most out of a prompt.
But that’s not all!
A new "Catch Up" chat interface is on its way. It will provide personal insights based on your recent activity, like “You have a meeting with the sales VP on Thursday. Let’s get you prepared — click here to get detailed notes.”
Plus, you’ll soon be able to create, publish, and manage prompts in Copilot Lab that are expressly tailored to your closest teams.
These features will be available to Copilot for Microsoft 365 users in the coming months.
But in the meantime, you should definitely check Microsoft’s joint report with LinkedIn on the state of AI at work to see just how much workers are embracing AI to improve productivity.
Trust me, you don't want to be left behind!
🤩 OpenAI is Developing New AI Detection Tools!
OpenAI has been working on detecting AI-generated content for years, and they’re not showing signs of slowing down!
Earlier this year, they added C2PA watermarks to images generated on the ChatGPT website and the API for the DALL-E 3 model.
And now, they're developing new provenance methods to track content and prove whether it's AI-generated.
These methods include an “image detection classifier” that uses AI to identify if a photo was created by DALL-E 3 and a “tamper-resistant watermark” that can tag content like audio with invisible signals.
OpenAI claims the classifier is so impressive that it can accurately detect DALL-E 3 generated images with about 98% accuracy, even if the image is cropped, compressed, or changes in saturation.
But there’s a catch — the tool’s performance in determining whether the content was from other AI models is not as good.
It flagged only 5 to 10 percent of pictures from other image generators, like Midjourney.
However, both the image classifier and audio watermarking are still a work in progress, so there's hope for improvement.🤞
And to achieve this, OpenAI says they’ll need feedback from users!
They’re even inviting researchers and nonprofit journalism groups to test the tools via their research access platform so they can identify areas for improvement.
Are you interested? Then click here for more info!
🧱Around The AI Block
🚀 4 best AI apps you need for better work productivity.
Acrobat AI Assistant: Interact with your document for quick answers and one-click summaries.
Web3 Summary: Say goodbye to the headaches of decoding crypto using AI.
Argil: Create shorts of yourself in under 5 minutes with your AI clone.
🤖ChatGPT Prompt Of The Day: Simplify complex sentences
Simple, straightforward language tends to be more engaging and easier to follow.
It keeps your audience interested and focused on your message rather than struggling to understand complex sentences.
Check out this ChatGPT prompt to help you with that.
I want you to help me simplify the following sentences: “”, which is from , written in . Your task is to break it down into smaller, more easily understood parts and write it in a more tone. Your goal is to ensure the simplified version is more accessible to .
