Elon Musk’s startup xAI premiered a beta version of its artificial intelligence assistant known as Grok-2, the company’s latest large language model, on Wednesday, as the company races to catch up against OpenAI’s ChatGPT tool.
Like OpenAI’s DALL-E, Google’s Gemini and Midjourney, the Grok-2 AI model includes an image generation tool; however, unlike those competitors, Musk’s version has fewer restrictions on what kinds of images can be generated, according to Forbes. The beta version is accessible to X users who pay for a "Premium" or "Premium+" subscription.
In its announcement, xAI said Grok-2 "is our state-of-the-art AI assistant with advanced capabilities in both text and vision understanding, integrating real-time information from the X platform, accessible through the Grok tab in the X app." The company also launched Grok-2 mini, a "small but capable model that offers a balance between speed and answer quality."
"Compared to its predecessor, Grok-2 is more intuitive, steerable, and versatile across a wide range of tasks, whether you're seeking answers, collaborating on writing, or solving coding tasks," the start-up said in a statement. "In collaboration with Black Forest Labs, we are experimenting with their FLUX.1 model to expand Grok’s capabilities on X."
Independent AI benchmark sites have already ranked Grok-2 among the top five chatbots across the world, according to the Financial Times.
Musk on Wednesday re-posted an image of George Washington created by the Grok-2 AI model.
"Maybe there is hope for humanity," X user Mario Nawfal wrote to his 1.6 million followers.
Google’s Gemini was slammed for generating AI images of America’s Founding Fathers as Black women.
Musk also shared an AI-generated video of him and former President Trump dancing to "Staying Alive."
"Haters will say this is AI," he quipped.
He added in another post, "Subscribe to X Premium for the most fun!"
Musk co-founded OpenAI in 2015 with its current CEO Sam Altman and Greg Brockman. He parted ways with his fellow co-founders three years later following a disagreement over the direction of OpenAI and their plan to turn the non-profit into a for-profit company.
In a new lawsuit this month in California federal court, Musk accuses OpenAI and Altman of "preying on Musk’s humanitarian concern about the existential dangers posed by artificial intelligence" and having "assiduously manipulated" Musk into investing in and co-founding the non-profit venture "by promising that it would charter a safer, more open course than profit-driven tech giants."
The lawsuit says Musk was promised OpenAI would serve as a "meaningful counterweight to Google’s DeepMind" and Altman conned him into believing the start-up would "focus on safety and openness for the benefit of humanity, not shareholder value." OpenAI, backed by Microsoft, dismissed the allegations as "incoherent and frivolous."
AI ADVANCEMENTS CAN BE BOTH A TOOL AND A THREAT, CYBERSECURITY OFFICIALS SAY
OpenAI network’s valuation has since soared to $100 billion, Musk’s lawsuit says.
Musk, who’s accused ChatGPT of being biased, politically correct and "woke," later launched xAI in March 2023.
Since then, xAI has accrued a valuation of $18 billion and secured $6 billion in Series B funding in May. Among the investors are Valor Equity Partners, Vy Capital, Andreessen Horowitz, Sequoia Capital, Fidelity Management & Research Company, Prince Alwaleed Bin Talal and Kingdom Holding.
Musk recently said he is vying for approval from Tesla’s board to invest another $5 billion into xAI, the Financial Times reported.
That would bring xAI’s investment close to the same as OpenAI’s $13 billion and outpace the $9 billion in investments received by Anthropic.
LMSYS, a leading site for comparing or benchmarking AI model capabilities, ranked Grok-2’s performance as significantly better than Meta’s and Anthropic’s best models, the Financial Times reported. OpenAI’s latest model, GPT-4o, propelled it back to first place in performance, with Google’s Gemini Pro in second.
While working to improve the Grok-1.5 model, which experts criticized for "hallucinations," xAI said developers focused on "following instructions and providing accurate, factual information."
The company said Grok-2 "has shown significant improvements in reasoning with retrieved content and in its tool use capabilities, such as correctly identifying missing information, reasoning through sequences of events, and discarding irrelevant posts."
