Topics
late
AI
Amazon
Image Credits:Gabby Jones/Bloomberg / Getty Images
Apps
Biotech & Health
Climate
Image Credits:Gabby Jones/Bloomberg / Getty Images
Cloud Computing
Commerce
Crypto
initiative
EVs
Fintech
Fundraising
Gadgets
Gaming
Government & Policy
Hardware
layoff
Media & Entertainment
Meta
Microsoft
seclusion
Robotics
Security
societal
infinite
startup
TikTok
transferral
Venture
More from TechCrunch
event
Startup Battlefield
StrictlyVC
Podcasts
Videos
Partner Content
TechCrunch Brand Studio
Crunchboard
Contact Us
Elon Musk - owned Xlaunched Grok-2 and Grok-2 miniin beta today with improved reasoning . The unexampled Grok AI model can now generate images on the X societal meshwork , though Grok get at is presently limited to Premium and Premium+ users on X.
According to xAI , both Grok-2 and Grok-2 mini are available to user on X in genus Beta .
“ We are worked up to release an early preview of Grok-2 , a pregnant stone’s throw forward from our late model Grok-1.5 , featuring frontier capableness in chat , coding , and reasoning . At the same time , we are introducing Grok-2 miniskirt , a diminished but capable sib of Grok-2 . An early version of Grok-2 has been prove on the LMSYS leaderboard under the name ‘ sus - column - r , ’ ” xAI ’s blog post about Grok-2 read .
Musk ’s AI fellowship is also planning to make both role model available to developers through its enterprise API later on this month .
other persona get by user suggest that Grok ’s mental image - generation feature of speech does n’t have any guardrail around creating images of political figures — and many users are currently taking advantage of this . However , with the U.S. presidential election around the corner , the company will likely confront pressure to add limits on these features .
Grok 2.0 … . Ohh boyyyy 😆 😆 😆 pic.twitter.com/TjzB7WMhVp
Grok 2.0 will do political illustration and real people , while ChatGPT refuses . This directly make Grok 10x more fun……pic.twitter.com/yDBJO0jWba
Grok AI Image Generation : pic.twitter.com / mjzSMEJ5Df
The text below the sample image prompts suggests that Grok is using FLUX.1 by Black Forest Labs to make these images .
There are few particular on Grok 2 ’s genuine capabilities at the second . App researcher ( and apparently ten characteristic examiner ) Nima Owji claim that Grok 2 is secure at codification coevals , writing , and newsworthiness , though there is n’t much to plump for that . Notably , Grok ’s first version has slipped upmany timeswhilesummarizing news program .
Join us at TechCrunch Sessions: AI
Exhibit at TechCrunch Sessions: AI
BREAKING : Here ’s an early look at Grok 2.0 characteristic and abilities!It ’s skilful at cipher , writing , and generating news ! It ’ll also generate images using the FLUX.1 model!pic.twitter.com / UlDW2Spen8
Without any demarcation line on mental image genesis , Grok could well become a cock to create misinformation on X and other platform . It ’s also not clean-cut if Grok - powered images embed any metadata to indicate that they are AI sire .
We have asked X about how it plans to limit double generation for harmful purposes and will update the story if we get a line back . However , the company has seldom reply to media question since Musk ’s takeover .
xAI note that the company plans to deploy Grok-2 and Grok-2 mini in AI - driven features on X , including improved hunt capabilities , post analytics , and reply procedure . This could have in mind that X might introduce AI - power replies . The society also say that it will bring out a preview of multimodal intellect as a core part of Grok ’s experience on X and the API .
