

BREAKING : Google may have started a rollout for its Canvas feature on Gemini to some users. So far, I've only seen one report from Japan. The Canvas feature will allow users to work with documents and code straight inside Gemini's response.

By
–


BREAKING : Google may have started a rollout for its Canvas feature on Gemini to some users. So far, I've only seen one report from Japan. The Canvas feature will allow users to work with documents and code straight inside Gemini's response.
By
–
Tesla has been a trailblazer in EVs, but 2025 has been turbulent—global sales have plummeted, and competition is rising. Some blame Elon Musk’s political ties and controversial comments, while others point to cheaper, more advanced rivals like #BYD. So, what do you think?

By
–
Figure announced the launch of BotQ, a new automation-driven manufacturing facility
— Rowan Cheung (@rowancheung) 18 mars 2025
It can churn out 12,000 humanoids a year—and will be scaled to build a fleet of 100,000 humanoids within four yearspic.twitter.com/lgTkq9HtQ2
Figure announced the launch of BotQ, a new automation-driven manufacturing facility It can churn out 12,000 humanoids a year—and will be scaled to build a fleet of 100,000 humanoids within four years

By
–
Chinese giant Baidu released two aggressively priced multimodal AI models
— Rowan Cheung (@rowancheung) 18 mars 2025
— ERNIE 4.5 with enhanced EQ outperforms GPT-4.5 at just 1% of its price
—ERNIE X1 deep thinking model matches DeepSeek's R1 at half the price
China continues to cook!pic.twitter.com/FSrmEgiF8r
Chinese giant Baidu released two aggressively priced multimodal AI models — ERNIE 4.5 with enhanced EQ outperforms GPT-4.5 at just 1% of its price
—ERNIE X1 deep thinking model matches DeepSeek's R1 at half the price China continues to cook!

By
–
Perplexity signed an agreement with SoftBank to grow Perplexity Enterprise Pro in Japan
— Rowan Cheung (@rowancheung) 18 mars 2025
SoftBank will deploy its 7,000-member sales team to accelerate Perplexity's adoption within businesses.
They have also been using Perplexity internally!pic.twitter.com/IS4AeyXVow
Perplexity signed an agreement with SoftBank to grow Perplexity Enterprise Pro in Japan SoftBank will deploy its 7,000-member sales team to accelerate Perplexity's adoption within businesses. They have also been using Perplexity internally!

By
–
Tencent dropped two new versions of its Hunyuan 3D generation AI
— Rowan Cheung (@rowancheung) 18 mars 2025
—3D 2.0 MV fine-tuned from Hunyuan3D-2 to support multiview shape generation.
— 3D 2.0 Mini contains a 0.6B shape generator, which is smaller and faster than the previous 1.1B onepic.twitter.com/8cYsUIh9sv
Tencent dropped two new versions of its Hunyuan 3D generation AI —3D 2.0 MV fine-tuned from Hunyuan3D-2 to support multiview shape generation.
— 3D 2.0 Mini contains a 0.6B shape generator, which is smaller and faster than the previous 1.1B one

By
–
Zoom's AI Companion is going agentic with new tools to identify and complete tasks across the platform It is getting new memory and reasoning smarts to leverage the right agentic tools for each task, like executing action items or scheduling follow-ups

By
–
xAI acquired Hotshot, a startup developing AI for generating videos in various styles
— Rowan Cheung (@rowancheung) 18 mars 2025
Terms of the deal remain undisclosed, although founder Aakash says they'll continue training on xAI's Colossus cluster
Expect strong video generation from Grok soon!pic.twitter.com/epEc60n46d
xAI acquired Hotshot, a startup developing AI for generating videos in various styles Terms of the deal remain undisclosed, although founder Aakash says they'll continue training on xAI's Colossus cluster Expect strong video generation from Grok soon!
By
–
TODAY'S AI NEWS: Roblox just made 3D content creation as simple as a text command! Plus, more news from Mistral, xAI, Zoom, Tencent, Perplexity, Baidu, and more. Here's everything you need to know:

By
–
At NVIDIA GTC 2025, Li Auto, a Chinese car company, released more details about MindVLA, their next-generation autonomous driving model. MindVLA is a world model that deeply integrates their Novel View Restoration and Progressive Scene Reconstruction capabilities.