Elon Musk unveiled Grok 3, the latest iteration of xAI’s artificial intelligence chatbot, claiming it represents “an order of magnitude more capable” than its predecessor. During a Monday livestream on X (formerly Twitter), Musk asserted that Grok 3 outperforms OpenAI’s GPT-4o and competing systems from Google, DeepSeek, and Anthropic across critical benchmarks in mathematics, science, and coding.
The launch has generated mixed reactions from AI researchers and industry experts who gained early access to the model. Andrej Karpathy, OpenAI cofounder and former Tesla AI lead, acknowledged that Grok 3 successfully handled complex information retrieval tasks but struggled with certain “tricky” tic-tac-toe problems, similar to OpenAI’s o1-pro model. Karpathy estimated Grok 3’s capabilities fall “somewhere around o1-pro capability, and ahead of DeepSeek-R1,” though he emphasized the need for comprehensive evaluations.
xAI also introduced DeepSearch, a new tool designed to enhance research, brainstorming, and data analysis while providing transparency about its query processing methods. However, Karpathy noted that DeepSearch currently resembles Perplexity’s offering but hasn’t yet matched the thoroughness of OpenAI’s Deep Research tool.
Ethan Mollick, a Wharton School professor specializing in AI research, characterized Grok 3 as “a very solid frontier model” but not compelling enough to switch from existing options. He observed that while Grok 3 excels in select benchmarks against some OpenAI models, it doesn’t clearly surpass OpenAI’s o3 system.
Gary Marcus, founder of Geometric Intelligence and prominent AI skeptic, delivered harsher criticism, stating Musk’s promise of “the smartest AI ever” fell short. Marcus described the launch as lacking any “game changer” compared to OpenAI’s offerings.
The release intensifies the rivalry between Musk and OpenAI CEO Sam Altman, which escalated after Musk’s recent $97.4 billion unsolicited bid to acquire OpenAI—promptly rejected by the company’s board. Musk revealed that Grok 3 utilizes “more than 10 times” the computational power of its predecessor and completed pre-training in early January. The model is immediately available to X Premium+ subscribers, with xAI launching a new SuperGrok subscription service for mobile and web users. Musk indicated plans to open-source previous model versions once Grok 3 reaches maturity, estimated within “a few months.”
Key Quotes
We’re continually improving the models every day, and literally within 24 hours, you’ll see improvements
Elon Musk made this statement during the Grok 3 livestream presentation on X, emphasizing xAI’s commitment to rapid iteration and continuous model enhancement, suggesting an aggressive development timeline compared to competitors.
The impression overall I got here is that this is somewhere around o1-pro capability, and ahead of DeepSeek-R1, though of course we need actual, real evaluations to look at
Andrej Karpathy, OpenAI cofounder and former Tesla AI director, provided this measured assessment after early testing of Grok 3, positioning it as competitive but not definitively superior to existing frontier models.
A very solid frontier model, but not one you would stop using your current frontier model for
Ethan Mollick, Wharton School professor and AI researcher, offered this pragmatic evaluation on LinkedIn, suggesting that while Grok 3 is technically impressive, it doesn’t provide sufficient differentiation to justify switching from established alternatives.
Sam Altman can breathe easy for now. No major leap forward here
Gary Marcus, AI critic and Geometric Intelligence founder, delivered this pointed assessment on Substack, dismissing Musk’s claims of revolutionary advancement and suggesting OpenAI maintains its competitive position despite xAI’s efforts.
Our Take
The Grok 3 launch reveals the complex dynamics shaping today’s AI landscape. Despite Musk’s bold claims, the tepid expert response suggests we’re entering a phase where incremental improvements dominate over breakthrough innovations. The fact that respected researchers like Karpathy see Grok 3 as roughly equivalent to existing models indicates the frontier of AI capabilities may be consolidating rather than rapidly expanding.
What’s particularly noteworthy is how personal rivalries are driving corporate strategy in AI development. Musk’s aggressive timeline and public positioning against OpenAI appear motivated as much by his conflict with Altman as by pure technical ambition. This raises questions about whether competitive dynamics are optimizing for genuine innovation or merely spectacle.
The mixed reception also highlights a maturation in AI evaluation—experts are moving beyond marketing claims to demand rigorous, independent testing. This skepticism is healthy for the industry, potentially tempering unrealistic expectations while focusing development on practical, measurable improvements that deliver real value to users and enterprises.
Why This Matters
Grok 3’s launch represents a significant escalation in the AI arms race, particularly between Elon Musk’s xAI and OpenAI, the company he cofounded but later departed. This development underscores the intensifying competition among tech giants to dominate the generative AI market, which analysts project will reach hundreds of billions in value over the coming years.
The mixed expert reactions highlight a crucial industry reality: achieving meaningful performance leaps in frontier AI models is becoming increasingly difficult. While Grok 3 demonstrates competitive capabilities, the lack of clear superiority suggests the field may be approaching certain technical plateaus, requiring breakthrough innovations rather than incremental improvements.
For businesses and enterprises, this competition drives rapid innovation and potentially lower costs as companies vie for market share. However, the fragmented landscape also creates challenges in selecting and integrating AI solutions. The personal rivalry between Musk and Altman adds an unpredictable element to the industry’s trajectory, potentially influencing investment decisions, talent acquisition, and strategic partnerships. As AI becomes increasingly central to business operations and societal functions, the concentration of power among a few competing entities raises important questions about access, governance, and the future direction of artificial intelligence development.
Recommended Reading
For those interested in learning more about artificial intelligence, machine learning, and effective AI communication, here are some excellent resources:
Recommended Reading
Related Stories
- Google’s Gemini: A Potential Game-Changer in the AI Race
- OpenAI CEO Sam Altman’s Predictions on How AI Could Change the World by 2025
- Elon Musk’s ‘X’ AI Company Raises $370 Million in Funding Round Led by Himself
- Elon Musk Drops Lawsuit Against ChatGPT Maker OpenAI, No Explanation
- Mistral AI’s Consumer and Enterprise Chatbot Strategy
Source: https://www.businessinsider.com/grok-3-openai-chatgpt-elon-musk-xai-sam-altman-2025-2