Grok-3 Launch: Musk's xAI unveils its most advanced AI model yet
Pranjal Chandra | Feb 19, 2025, 15:46 IST
( Image credit : ETtech )
Elon Musk’s AI startup, xAI, has launched Grok-3, a new AI chatbot, claiming it outperforms Google, Anthropic, and Meta's models. This positions xAI against OpenAI and DeepSeek. Additionally, xAI introduced 'Deep Search,' an AI-powered search engine aimed at revolutionizing online research.
Elon Musk’s artificial intelligence startup, xAI, has unveiled Grok-3, the latest iteration of its AI chatbot, with bold claims of outperforming rivals from Google, Anthropic, and Meta. This release signals a new chapter in the competitive AI landscape, where Musk is positioning xAI as a major player against OpenAI and DeepSeek.
Grok-3: A leap forward in
According to xAI, Grok-3 represents a significant leap from its predecessor, Grok-2, boasting ten times more computing power. The model’s rollout began on Monday evening, with early access granted to X’s Premium+ subscribers. Additionally, xAI introduced “SuperGrok,” a new subscription tier that promises exclusive access to the AI’s most advanced features.
Musk, during a livestreamed demo on X, highlighted Grok-3’s capabilities, stating that it is “an order of magnitude more capable than Grok 2 in a very short period of time.” He emphasized that the model remains in beta, with continuous improvements expected on a daily basis. Among the upcoming features is voice interaction, slated for release within a week.
xAI developers shared benchmark data showcasing Grok-3’s strengths in mathematical reasoning, science, and coding. According to the Artificial Analysis Quality Index—an independent AI ranking system—Grok-3 surpasses models from Google, Anthropic, and Meta but trails behind OpenAI’s latest iterations, o3 and o1, as well as DeepSeek-R1. However, the Grok-3 Reasoning Beta variant ranks second only to OpenAI’s o3 in reasoning performance.
While xAI touts Grok-3’s advancements, industry experts remain cautious. Former Tesla AI director Andrej Karpathy, a key figure in Musk’s AI ventures, praised the model’s benchmark results as “quite encouraging.” However, some AI enthusiasts on X expressed skepticism, waiting for real-world testing to validate xAI’s claims.
Beyond chatbots: xAI introduces
Alongside Grok-3, xAI announced “Deep Search,” a new AI-powered search engine designed to rival offerings from OpenAI and Perplexity. Musk described Deep Search as a tool capable of significantly accelerating online research, stating, “Something that might take you half an hour or an hour of researching on the web or searching social media, you can just ask it to go do that and come back, and 10 minutes later, it’s done an hour’s worth of work for you.”
This move signals Musk’s ambition to redefine the AI-powered information retrieval landscape, offering a direct challenge to traditional search engines and AI-driven search platforms.
Musk has long positioned Grok as a maverick AI, aiming to challenge what he sees as the “woke” biases of mainstream AI models. During Monday’s stream, he reiterated that Grok is designed to be a “maximally truth-seeking AI, even if that truth is sometimes at odds with what is politically correct.”
Despite this branding, early users found that Grok-3’s responses often align with conventional AI models, maintaining similar ethical guardrails. However, controversy arose when Musk shared a screenshot of Grok-3 allegedly criticizing tech news outlet The Information while endorsing X as the ultimate source of “real, trustworthy news.”
Following Musk’s post, some observers speculated that xAI may be fine-tuning Grok-3 to reflect his personal views. However, tests conducted by NBC News showed different results, with Grok-3 offering a neutral stance on The Information, describing it as a “well-regarded tech news outlet known for its in-depth reporting and analysis.”
This discrepancy raises questions about how xAI fine-tunes Grok-3’s responses. AI models often generate different answers depending on context, phrasing, and real-time updates. Whether xAI is deliberately shaping Grok-3’s tone or whether it naturally reflects Musk’s perspective remains a subject of debate.
A growing
The launch of Grok-3 is the latest development in Musk’s ongoing rivalry with OpenAI, the company he co-founded but later distanced himself from. As OpenAI continues to push the boundaries with its cutting-edge models, Musk is determined to establish xAI as a formidable competitor.
With Grok-3’s launch, xAI is signaling its intent to challenge the AI industry’s biggest players. However, whether it can live up to the hype remains to be seen as real-world testing and user feedback unfold.
As the AI race intensifies, Grok-3’s performance—and xAI’s long-term strategy—will determine whether Musk’s vision of a truth-seeking AI can truly disrupt the status quo or whether it will remain an ambitious, yet polarizing, experiment.
Grok-3: A leap forward in AI capabilities
Musk, during a livestreamed demo on X, highlighted Grok-3’s capabilities, stating that it is “an order of magnitude more capable than Grok 2 in a very short period of time.” He emphasized that the model remains in beta, with continuous improvements expected on a daily basis. Among the upcoming features is voice interaction, slated for release within a week.
Performance benchmarks: how Grok-3 stacks up
While xAI touts Grok-3’s advancements, industry experts remain cautious. Former Tesla AI director Andrej Karpathy, a key figure in Musk’s AI ventures, praised the model’s benchmark results as “quite encouraging.” However, some AI enthusiasts on X expressed skepticism, waiting for real-world testing to validate xAI’s claims.
Beyond chatbots: xAI introduces deep search
This move signals Musk’s ambition to redefine the AI-powered information retrieval landscape, offering a direct challenge to traditional search engines and AI-driven search platforms.
The Grok identity: “Maximally Truth-Seeking” AI
Despite this branding, early users found that Grok-3’s responses often align with conventional AI models, maintaining similar ethical guardrails. However, controversy arose when Musk shared a screenshot of Grok-3 allegedly criticizing tech news outlet The Information while endorsing X as the ultimate source of “real, trustworthy news.”
AI or algorithmic alignment? the question of bias
This discrepancy raises questions about how xAI fine-tunes Grok-3’s responses. AI models often generate different answers depending on context, phrasing, and real-time updates. Whether xAI is deliberately shaping Grok-3’s tone or whether it naturally reflects Musk’s perspective remains a subject of debate.
A growing AI Rivalry : Musk vs. OpenAI
With Grok-3’s launch, xAI is signaling its intent to challenge the AI industry’s biggest players. However, whether it can live up to the hype remains to be seen as real-world testing and user feedback unfold.
As the AI race intensifies, Grok-3’s performance—and xAI’s long-term strategy—will determine whether Musk’s vision of a truth-seeking AI can truly disrupt the status quo or whether it will remain an ambitious, yet polarizing, experiment.