Hello, AI Lovers!

Your daily AI brief is here to kickstart your day with the top stories you need to know.

In today’s AI Shorts:

• OpenAI’s Voice Engine Still on Hold & other top stories

• Learn ChatGPT – Automate Data Entry from PDFs with ChatGPT

• Learn to Turn sales email creation into an automated process!

• Learn to Discover 5 new AI tools

• Other top stories in AI

• Opinions on AI across the web

Let’s dive in and get you up to speed.

TOP STORIES

1. OpenAI Voice Cloning Tool Faces Delays Amid Safety Concerns

   OpenAI’s Voice Engine, announced in 2023, remains unreleased to the public due to concerns over misuse and regulatory scrutiny. Initially scheduled for a March 2024 launch, only 10 developers have access, focusing on ethical AI deployment. Despite successful tests, OpenAI highlights challenges like election security and fraud prevention. Developers need consent for voice use, and safeguards include watermarking and potential voice authentication.

   Read in detail here.

2. Former OpenAI Policy Lead Criticizes Revised AI Safety Narrative

   A former OpenAI policy lead, Miles Brundage, criticizes the company for altering its AI safety narrative, particularly regarding the cautious deployment of AI models like GPT-2. He argues that OpenAI’s current approach misrepresents its history, potentially downplaying safety concerns in the race for AGI development. This criticism raises questions about OpenAI’s commitment to responsible AI advancements and long-term safety.

   Read in detail here.

3. Alibaba’s AI Model QwQ-32B Rivals Giants with Efficiency

   Alibaba’s QwQ-32B AI model, built on the Qwen2.5-32B foundation, matches performance of larger competitors like DeepSeek R1, boasting 32.5 billion parameters. QwQ-32B excels in mathematical reasoning and coding, achieving competitive results. The model signifies a shift in the AI industry, emphasizing efficient training over raw parameter count and advancing towards Artificial General Intelligence (AGI).

   Read in detail here.

4. DuckDuckGo Enhances AI Search Interface

   DuckDuckGo is advancing its AI capabilities, rebranding its chatbot interface as Duck.ai and exiting beta. Duck.ai uses models from Anthropic, OpenAI, and Meta for conversational search results. While the service remains free with query limits, a paid plan is being considered. DuckDuckGo emphasizes privacy, with anonymized chats and options for AI-generated answers. Enhanced integration with conventional search aims to balance user preferences and AI assistance.

   Read in detail here.

AI MASTERY

CHATGPT MASTERY

Automate Data Entry from PDFs with ChatGPT

Did you know ChatGPT has a project feature that can revolutionize your data entry workflow? Imagine turning those stacks of PDFs into perfectly formatted spreadsheets with just a few clicks. With ChatGPT’s Custom Instructions, you can now automate data extraction and conversion, eliminating the need for tedious manual entry.

HOW TO SET IT UP:

1. Access ChatGPT: Open ChatGPT in your web browser.

2. Set Custom Instructions:

   • Click on the “Add Instructions” button.

   • Part 1: Inform ChatGPT that you are converting from PDF to CSV format.

   • Part 2: Specify the desired format of the CSV, including column names and data types.

   • Save the instructions.

3. Attach PDF Files: Upload the PDF files containing the data you want to convert.

4. Submit Without Prompt: Submit the files without adding a specific prompt; ChatGPT will follow the custom instructions.

5. Review and Download: Check the table format response from ChatGPT, ensuring it matches your specified columns.

6. Download CSV: Click the download button (usually in the top right corner) to save the data as a CSV file.

Why It’s Useful: Custom Instructions automate tedious data entry, converting unstructured PDF data into structured CSV files, saving you hours of manual work.

Pro Tip: Refine your custom instructions based on the complexity of the PDF and the specific data you need. Clear and detailed instructions will yield more accurate and consistent results.

AI TUTORIAL

Turn sales email creation into an automated process!

Regie.ai automates sales email writing and sequence creation, helping teams craft personalized, high-converting emails. It optimizes outreach, streamlines prospecting, and uses AI insights to boost engagement and save time.

Step-by-step:

1. Log into Regie.ai: Access the Regie.ai website and log into your account to get started. If you’re new, sign up to begin automating your sales emails.

2. Navigate to Sequence Creation: Find the option to create a new email sequence in the interface. Look for labels like “New Sequence” or “Create Campaign” to begin.

3. Define Your Target Audience: Specify your ideal customer’s characteristics for tailored emails. Provide detailed information to help Regie.ai personalize effectively.

4. Choose Your Sequence Type: Decide whether to start from scratch or use a pre-designed template. Templates save time, while starting fresh allows customization.

5. Input Key Information: Enter essential details like the product, pain points, and desired call to action. This will guide Regie.ai in generating relevant content.

6. Generate Email Content: Use Regie.ai’s AI writing tools to create email content for each step. Specify tone, length, and purpose for each email in the sequence.

7. Personalize and Refine: Review and personalize the generated content to make it more engaging. Add specific details that resonate with your target audience for better results.

8. Schedule and Activate: Set the timing and frequency of each email and activate the sequence. Start sending emails automatically once you’re satisfied with the setup.

Pro tip: Monitor sequence performance using Regie.ai’s analytics and fine-tune your approach. Experiment with subject lines, content, and timing to optimize engagement and conversion rates.

DISCOVER NEW AI TOOLS

1. Supermanage – AI-driven meeting assistant that scans Slack messages and generates intelligent meeting briefs, ensuring managers walk into meetings fully prepared. Ideal for remote teams and busy executives.

2. Dumme – AI-powered tool that automatically extracts and edits short-form video clips from long videos, optimizing them for TikTok, Instagram Reels, and YouTube Shorts. A must-have for content creators looking to repurpose long-form content.

3. Delve AI – AI-driven customer research tool that creates automated buyer personas by analyzing website traffic and customer data. Great for marketers, sales teams, and businesses optimizing their audience targeting.

4. Unbound – AI-powered productivity suite that automates content creation, marketing copy, and image generation. Ideal for startups and solopreneurs looking to scale their content efficiently.

5. Lime AI – AI-driven legal assistant that automates contract analysis, compliance checks, and legal research. Perfect for lawyers, businesses, and compliance officers needing quick insights.

OTHER STORIES

OTHER TOP STORIES

1. Larry Page’s New AI Startup Dynatomics

   Larry Page is reportedly launching Dynatomics, an AI-driven company focused on creating optimized designs for manufacturing. The startup, led by ex-Kittyhawk CTO Chris Anderson, aims to revolutionize product creation with advanced AI technology.

   Read in detail here.

2. Hugging Face CSO Raises AI Concerns

   Hugging Face’s Thomas Wolf warns current AI lacks true creativity, becoming “yes-men on servers.” He stresses the need for AI to challenge existing knowledge, ask novel questions, and avoid merely reiterating known answers.

   Read in detail here.

3. Google’s AI Search Expansion

   Google expands AI search results with Gemini 2.0, enhancing complex queries and multimodal searches. AI Mode, a chatbot-like feature, is in testing. Critics highlight accuracy issues, but Google pushes forward with AI integration across services.

   Read in detail here.

4. AI-Powered Cyber Threats Rising

   Darktrace report reveals 78% of cybersecurity leaders view AI-driven threats as significant. Companies face increased cybercrime-as-a-service and automation, yet 60% feel prepared. The skills gap and talent shortage remain critical challenges.

   Read in detail here.

5. Anthropic’s Claude Code Tool Bug Causes System Issues

   Anthropic’s Claude Code tool launch faced challenges with buggy auto-update commands, leading to unstable and “bricked” systems. The commands affected critical system files’ access permissions, prompting Anthropic to remove the issues and provide a troubleshooting guide.

   Read in detail here.

6. Nvidia AI Server Maker Hon Hai’s Revenue Surges

   Hon Hai Precision Industry Co., Nvidia’s AI server supplier, reports a 25% revenue increase in early 2025, reaching NT$1.1 trillion. This growth reflects rising AI computing demand and outpaces last year’s 11% increase.

   Read in detail here.

7. Scale AI Investigated for Labor Violations

   The U.S. Department of Labor investigates Scale AI for allegedly underpaying and mistreating data labeling contractors. Scale AI disputes claims, emphasizing compliance with labor laws, fair pay, and timely payments while collaborating with the department to clarify its practices.

   Read in detail here.

8. Musk’s AI Chatbot Grok Suggests Trump Ties to Putin

   Elon Musk’s AI chatbot, Grok, estimates a 75-85% likelihood that Trump is compromised by Putin. Factors include Trump’s financial ties to Russia, refusal to criticize Putin, and historical behavior patterns.

   Read in detail here.

9. Microsoft to Invest $300 Million in South Africa’s AI Infrastructure

   Microsoft will invest an additional $296.81 million in South Africa’s AI and cloud infrastructure by 2027, expanding its Azure services. The investment includes technical certification exams for 50,000 individuals in high-demand digital skills.

   Read in detail here.

10. Taco Bell Expands AI in Drive-Throughs

    US fast-food chain Taco Bell integrates AI to enhance drive-through efficiency and customer experience. The AI technology streamlines orders and reduces wait times, aiming to revolutionize quick-service restaurant operations.

    Read in detail here.

OPINIONS

1. Did Amazon Just Say Checkmate to OpenAI?

   Amazon’s Claude introduces hybrid reasoning, blending standard and extended thinking modes. This innovation enhances response depth and user engagement. While promising, experts debate its long-term impact on ChatGPT, as OpenAI aims to unify its model capabilities for market leadership.

   Read in detail here.

2. Can AI Plan a Better Overland Trip?

   While ChatGPT offers helpful trip-planning advice and insights, it falls short in delivering accurate, real-time data and lacks the personal touch. Traditional methods like satellite imagery, travel communities, and human intuition remain invaluable for overland trip planning.

   Read in detail here.

3. Debate on AI’s Future Role in Space

   Will AI or humans colonize space? Some believe AI, unrestricted by human biological needs and capable of self-replication, could be the true spacefarers, ensuring survival beyond Earth. Others argue for caution in advancing AI to avoid potential loss of human control.

   Read in detail here.

4. Average People Can Contribute to AI Safety

   While most individuals feel powerless in AI development, they can still contribute by becoming informed, spreading the message, participating in forums, conducting AI evaluations, donating money, protesting, and avoiding harmful actions. Collective efforts can significantly impact AI safety.

   Read in detail here.