Group 1 - UBS reiterated its Buy rating and $42 price target on Box, Inc. following the company's BoxWorks event in San Francisco, where new generative AI products were unveiled [1] - The company introduced several new AI products including Box AI Search Agent, Research Agent, Extract, Automate, and enhancements to Box AI Studio, along with Box Shield Pro as an add-on product [1] - Most of the AI suite was bundled into the Enterprise Advanced package, which opens additional monetization opportunities for the company [1] Group 2 - Feedback on the AI portfolio was mixed, with some partners actively promoting the Enterprise Advanced package while several customers indicated they were holding off on upgrades due to budget constraints [2] - The SKU for the Enterprise Advanced package had been available for less than a year, making it too early to see meaningful impact on results [2]

Model Overview - GPT5 is a hybrid model with both thinking and non-thinking versions, replacing previous models like GPT-4.0, GPT-4.1, and GPT-4.5 [2] - The model excels in coding, math, writing, health, and visual perception, adapting its response speed based on the complexity of the task [3][6] - GPT5 has three versions: standard, mini, and nano, with the mini version handling queries when usage limits are reached [7] - It features a 400,000 token context window, improving understanding of spacing, typography, and whitespace [9] Performance Benchmarks - GPT5 Pro achieved 100% on the Amy 2025 benchmark [18] - In enterprise metadata extraction, GPT5 shows a 5% to 8% improvement over GPT4.1%, averaging 90% overall accuracy [13] - GPT5 is 45% less likely to contain factual errors than GPT4.0, and when thinking, 80% less likely than OpenAI's 03 [27] - On the SWEBench verified coding benchmark, GPT5 achieved 74.9% compared to 30% with GPT4.0 [25] Availability and Versions - GPT5 is available to all users, with Plus subscribers getting more usage and Pro subscribers accessing GPT5 Pro [5] - GPT5 Pro is designed for challenging tasks, utilizing scaled parallel test time compute [36][37] - Box AI Studio offers GPT5 for enterprise document Q&A and analysis, trusted by over 100,000 organizations [13][14] Safety and Reliability - GPT5 communicates more honestly about its capabilities, recognizing when tasks cannot be completed [28][29] - It employs safe completions, a new safety training method, to provide helpful answers while staying within safety boundaries [32] Coding Capabilities - GPT5 is an excellent coding model, particularly in complex front-end generation and debugging larger repositories [8]

Model Performance - xAI's Grok 3 is a top performing model for single & multi-doc Q&A accuracy [1] - Grok 3 performs 9% better than Grok 2 for data extraction from documents [1] - Grok 3 performs strongly alongside other leading models when tested on complex legal contracts and real-world scenarios [1] - Grok 3 excels at sophisticated analytical tasks involving multi-step reasoning, precise information retrieval, quantitative analysis, and criteria-based ranking [2] - Grok 3 successfully handles challenging use cases such as extracting economic data from tables, analyzing HR frameworks, and evaluating SEC filings [2] - Grok 3 provides strong contextual understanding, factual accuracy, and reliable performance across various document types [2] Potential and Refinement - Grok 3 offers potential for advanced analytical workflows [3] - There is still room for refinement in language precision and handling highly complex logic [3]