Google Photos is launching a new feature that lets you turn yourself into a meme using AI. The feature, called Me Meme, uses Google's Gemini AI technology to combine meme templates with photos of yourself to create personalized memes.

Here's how it works: you select a meme template or upload your own, tap "add photo" to choose an image of yourself, and hit "Generate." The AI creates a version of the meme with your face in it. If you don't like the first result, you can regenerate it. Once you're happy, save the image or share it directly to other platforms.

Google is calling this an experimental feature, which means the results might be weird sometimes. The company suggests using well-lit, focused, front-facing photos for the best results and says it's adding more templates over time.

On the surface, this seems silly, but there's strategy behind it. These types of fun AI tools give people a reason to open Google Photos regularly rather than heading to a competitor's app. Google also has good reason to believe people will use this - OpenAI's Sora app, which lets you create AI videos featuring yourself, shot to number one on the U.S. App Store after launch. People love seeing themselves in AI-generated content.

The feature is rolling out to U.S. users on iOS and Android over the coming weeks under the "Create" tab. It's another example of Google weaving AI into everyday products, making the technology feel accessible rather than intimidating. Sometimes the best application of cutting-edge technology is just making people laugh at pictures of themselves.

Tesla is now offering fully driverless robotaxi rides in Austin, Texas, with no human safety driver in the front seat. CEO Elon Musk announced the milestone on Thursday, congratulating the Tesla AI team and making a recruitment pitch for engineers interested in what he calls "real-world AI."

Tesla first launched robotaxis in Austin last June, but those rides had a safety operator in the front passenger seat ready to take over if needed. The company began testing without safety drivers in December, and now it's officially offering paying customers rides in vehicles with no human backup.

According to Tesla's AI lead Ashok Elluswamy, the company is starting with just a few unsupervised vehicles mixed into the broader robotaxi fleet that still has safety monitors. The ratio of driverless to monitored vehicles will increase over time. Reports from riders indicate Tesla is charging for the rides, and there appears to be a chase car following the driverless vehicles for now.

This puts Tesla in direct competition with companies like Waymo and Zoox, which have been operating driverless taxi services in various cities. However, those competitors typically didn't charge for rides immediately upon deploying fully autonomous vehicles, choosing instead to build confidence and gather data first.

The Austin deployment represents a significant milestone for Tesla, which has long promised self-driving capabilities but faced skepticism about when it would actually deliver. The company's approach differs from competitors in that it relies primarily on cameras and AI rather than the expensive lidar sensors that other autonomous vehicle companies use.

Elon Musk claim that this work will "likely lead to AGI" (artificial general intelligence) is characteristic hyperbole, but the Austin launch does represent meaningful progress in Tesla's autonomous driving ambitions.

Google is now offering free SAT practice exams powered by Gemini AI, potentially disrupting the expensive test prep industry. Students can simply prompt Gemini by typing "I want to take a practice SAT test" and the AI will provide a full practice exam, analyze the results, highlight strengths and weaknesses, and offer detailed explanations for incorrect answers.

The company partnered with education companies like Princeton Review to ensure the practice questions closely mirror what students will encounter on the actual SAT. Google frames this as a game-changer for students who can't afford personalized SAT tutoring, opening the door for more students to compete on equal footing.

But the move raises some thorny questions. For starters, it's a direct threat to human SAT tutors whose livelihoods depend on providing personalized coaching to college-bound students. With Google offering a free alternative powered by AI, the traditional tutoring industry could take a significant hit.

There are also broader concerns about AI's role in education. Teachers worry that students might lean too heavily on tools like Gemini and ChatGPT, potentially weakening their problem-solving and critical thinking skills. Some studies suggest that relying too much on AI can actually harm students' ability to tackle challenges independently.

This isn't Google's first push into AI-powered education. The company recently launched a Gemini feature that lets teachers create podcast-style audio lessons aimed at Gen Z students, along with tools that help teachers brainstorm ideas, build lesson plans, and tailor learning materials.

OpenAI is planning to ship its first hardware device in the second half of 2026, and recent reports suggest it could be a pair of AI-powered earbuds. The news comes from OpenAI's Chief Global Affairs Officer Chris Lehane, who confirmed the timeline at a panel in Davos.

The company created significant buzz last year when it acquired Jony Ive's startup for $6.5 billion, bringing the legendary Apple designer who created the iPhone and MacBook on board. CEO Sam Altman has described the potential device as more "peaceful and calm" than iPhones, and earlier reports indicated the company wanted to build something screen-free and pocketable.

According to reports from Asian publications and leakers, the device is codenamed "Sweet Pea" and will feature a unique design compared to existing earbuds. The earbuds could run on a custom 2-nanometer processor and handle AI tasks locally instead of relying on cloud processing. A Taiwanese newspaper reported that OpenAI is exploring manufacturing partnerships with either China's Luxshare or Taiwan's Foxconn, with ambitious first-year sales targets of 40 to 50 million units.

The strategic logic makes sense for OpenAI. While ChatGPT has nearly a billion weekly users, the company currently depends on other devices and platforms for distribution. Building its own hardware would give OpenAI more control over how people interact with its AI assistant and allow for exclusive, purpose-built features.

But replacing daily devices like AirPods won't be easy, especially without strong integration with existing operating systems like iOS and Android. And the track record for AI hardware hasn't been encouraging. Humane's AI Pin got sold to HP after flopping. Rabbit is still struggling after initial hype. The Friend AI companion necklace faced backlash for its marketing tactics.

That said, some AI wearables are gaining traction. Meta's Ray-Ban smart glasses have been improving to the point where Meta can't keep up with demand and had to pause international expansion. Amazon recently acquired Bee, an AI meeting recorder that could function as a companion device.

B1/1.3 sensor, Dual Native ISO, and AI noise reduction deliver crisp 4K video and pro audio with four smart pickup modes for streaming and meetings.