On Tuesday, Elon Musk’s AI firm xAI introduced the beta launch of two new language fashions, Grok-2 and Grok-2 mini, accessible to subscribers of his social media platform X (previously Twitter). The fashions are additionally linked to the lately launched Flux picture synthesis mannequin, which permits X customers to create largely uncensored photorealistic photographs that may be shared on the positioning.
“Flux, accessible via Grok, is a wonderful text-to-image generator, however it’s also actually good at creating pretend images of actual places and folks, and sending them proper to Twitter,” wrote frequent AI commentator Ethan Mollick on X. “Does anybody know if they’re watermarking these in any means? It could be a good suggestion.”
In a report posted earlier at the moment, The Verge famous that Grok’s picture era capabilities seem to have minimal safeguards, permitting customers to create probably controversial content material. Based on their testing, Grok produced photographs depicting political figures in compromising conditions, copyrighted characters, and scenes of violence when prompted.
The Verge discovered that whereas Grok claims to have sure limitations, comparable to avoiding pornographic or excessively violent content material, these guidelines appear inconsistent in observe. Not like different main AI picture turbines, Grok doesn’t seem to refuse prompts involving actual individuals or add figuring out watermarks to its outputs.
Given what individuals are producing up to now—together with photographs of Donald Trump and Kamala Harris kissing or giving a thumbs-up on the best way to the Twin Towers in an obvious 9/11 assault—the unrestricted outputs might not final for lengthy. However then once more, Elon Musk has made an enormous deal out of “freedom of speech” on his platform, so maybe the potential will stay (till somebody probably recordsdata a defamation or copyright go well with).
Folks utilizing Grok’s picture generator to shock brings up an previous query in AI at this level: Ought to misuse of an AI picture generator be the duty of the one that creates the immediate, the group that created the AI mannequin, or the platform that hosts the photographs? To date, there isn’t a clear consensus, and the scenario has but to be resolved legally, though a brand new proposed US regulation referred to as the NO FAKES act would presumably maintain X responsible for the creation of sensible picture deepfakes.
With Grok-2, the GPT-4 ceiling nonetheless holds
Wanting past photographs, in a launch weblog, xAI claims that Grok-2 and Grok-2 mini characterize important developments in capabilities, with Grok-2 supposedly outperforming some main rivals in latest benchmarks and what we name “vibemarks.” It is at all times clever to strategy these claims with a dose of skepticism, however it seems that the “GPT-4 class” of AI language fashions (these with related functionality to OpenAI’s mannequin) has grown bigger, however the GPT-4 barrier has not but been smashed.
“There are actually 5 GPT-4 class fashions: GPT-4o, Claude 3.5, Gemini 1.5, Llama 3.1, and now Grok 2,” wrote Ethan Mollick on X. “All the labs are saying there’s room left for continued big enhancements, however we haven’t seen any fashions actually leap above GPT-4… but.”
xAI says it lately launched an early model of Grok-2 to the LMSYS Chatbot Enviornment underneath the title “sus-column-r,” the place it reportedly achieved the next general Elo rating than fashions like Claude 3.5 Sonnet and GPT-4 Turbo. Chatbot Enviornment is a well-liked subjective vibemarking web site for AI fashions, nevertheless it’s been the topic of controversy lately when individuals disagreed with OpenAI’s GPT-4o mini putting so extremely within the rankings.
Based on xAI, each new Grok fashions present enhancements over predecessor Grok-1.5 in areas like graduate-level science information, normal information, and math problem-solving in benchmarks which have equally proved controversial. The corporate additionally highlighted Grok-2’s efficiency on visible duties, claiming state-of-the-art ends in visible math reasoning and document-based query answering.
The fashions are actually accessible to X Premium and Premium+ subscribers via an up to date app interface. Not like a few of its rivals within the open weights house, xAI is not releasing the mannequin weights for obtain or impartial verification. This closed strategy stands in stark distinction to latest strikes by Meta, which lately launched its Llama 3.1 405B mannequin for anybody to obtain and run regionally.
xAI plans to launch each fashions via an enterprise API later this month. The corporate says this API will function multi-region deployment choices and safety measures like necessary multifactor authentication. Particulars on pricing, utilization limits, or knowledge dealing with insurance policies haven’t but been introduced.
Photorealistic picture era apart, maybe Grok-2’s largest legal responsibility is its deep hyperlink to X, which provides it a bent to tug inaccurate data from tweets. It is a bit like should you had a buddy who insisted on checking the social media web site earlier than answering any of your questions, even when it wasn’t notably related.
As Mollick identified on X, this shut hyperlink will be annoying: “I solely have entry to Grok 2 mini proper now, and it looks like a stable mannequin, however usually appears ill-served by its RAG connection to Twitter,” he wrote. “The mannequin is fed outcomes from Twitter that appear irrelevant to the immediate, after which desperately tries to attach them into one thing coherent.”
