Welcome back to Daily Zaps, your regularly-scheduled dose of AI news ⚡️ 

Here’s what we got for ya today:

• 🚫 Microsoft reveals 40 jobs that AI might kill

• 🇺🇸 Federal chief artificial intelligence officer

• 🤓 ChatGPT’s new study mode

• 🎥 NotebookLM adds video overviews

Let’s get right into it!

Sign Up | Advertise | Tools

RESEARCH

Microsoft reveals 40 jobs that AI might kill

A Microsoft Research study has identified 40 jobs most exposed to advanced AI tools such as customer service, writing, journalism, translation, and data analysis due to significant overlap with AI capabilities like Copilot and ChatGPT. It also highlights 40 roles with minimal AI exposure, mainly those requiring physical or hands-on work.

The study cautions that high exposure doesn’t guarantee job loss, comparing the shift to ATMs, which ultimately increased bank teller roles by enabling them to focus on customer service. Instead, the findings suggest AI will reshape how work is done, enhancing productivity and efficiency, though critics warn this could still lead to job cuts as companies streamline operations.

40 jobs at higher risk from AI disruption

40 jobs considered safer from AI disruption

GOVERNMENT

Federal chief artificial intelligence officer

Gregory Barbaccia, the federal government’s chief information officer, has begun referring to himself as the Federal Chief Artificial Intelligence Officer (CAIO), a title not formally established by policy but emerging amid the Trump administration’s push for AI adoption. Barbaccia has used the dual titles in public remarks and social media, emphasizing AI’s role in augmenting—not replacing—human work to make government more efficient.

While neither the recent AI Action Plan nor related executive orders mention a federal CAIO role, Barbaccia appears to be leading both the CIO Council and a Chief AI Officers’ Council to coordinate AI efforts across agencies. The AI Action Plan calls for formalizing this interagency body, potentially positioning the CAIO role as influential in shaping federal AI strategy.

STARTUPS

ChatGPT’s new study mode

OpenAI has launched a new Study Mode for ChatGPT that acts as an interactive tutor, guiding students through problems step by step rather than providing direct answers. Designed to promote critical thinking and deeper learning, the mode uses Socratic questioning, hints, and self-reflection prompts tailored to a student’s skill level.

It supports homework, exam preparation, and exploring new concepts, offering a back-and-forth experience where correct answers advance the lesson and mistakes trigger explanations. While Study Mode doesn’t use OpenAI’s latest models and may show occasional inconsistencies, it was developed with input from educators and is currently free for all ChatGPT users, with Edu plan support coming soon.

BIG TECH

NotebookLM adds video overviews

Google has begun rolling out Video Overviews to NotebookLM as of July 29, allowing users to create visual summaries of information stored in their notebooks. Building on the existing Audio Overviews feature, Video Overviews let users select topics, define learning goals, and tailor videos to specific audiences, making complex data easier to understand.

The feature, currently available only in English, enhances NotebookLM’s usefulness for students and professionals by adding visual aids to AI-generated breakdowns. While the rollout may take a few days to reach all users, Google plans to expand language support and continue adding new AI-driven features to the platform.