Tag
Upwork integrates OpenAI's models across its platform to launch AI-powered features including Job Post Generator, Upwork Chat Pro, and Uma (Upwork's Mindful AI), improving user productivity and spending metrics.
Indeed deployed a fine-tuned smaller GPT model with OpenAI to deliver contextual job matching for millions of users, achieving 60% token efficiency improvement through fine-tuning and GPT-4-based data augmentation on dedicated instances.
OpenAI introduces Rule-Based Rewards (RBRs), a method to improve AI model safety by using explicit rules instead of human feedback in reinforcement learning. RBRs have been integrated into GPT-4 and subsequent models to maintain safety-helpfulness balance while reducing reliance on human feedback collection.
OpenAI introduced CriticGPT, a GPT-4-based model designed to catch errors in ChatGPT's code output. When human trainers use CriticGPT for code review, they outperform those without assistance 60% of the time, addressing a fundamental limitation of RLHF as models become increasingly capable.
Clay launched Claygent, an AI-powered web scraper using GPT-4 that enables intelligent data extraction and sales prospecting with 10x growth potential. The tool optimizes token usage through smart model selection and binary search approaches, with plans for vertical-specific fine-tuning and proactive trigger-based features.
OpenAI introduces sparse autoencoders as a method to extract and interpret concepts from large language models like GPT-4, addressing the fundamental challenge of understanding neural network behavior. They release a research paper, code, and feature visualization tools to help researchers train autoencoders at scale and improve AI safety through better interpretability.
MavenAGI launches an automated customer support platform powered by GPT-4 that processes customer data, integrates with enterprise systems, and uses self-evaluation to provide human-quality support responses. The system has been validated on over 1M customer interactions across Salesforce, Zendesk, and other platforms.
Canva's AI-powered Magic Studio has been used 5 billion times, leveraging OpenAI's GPT-4 to integrate generative AI tools for text, image, and video creation while maintaining an intuitive user experience for 175 million monthly users.
OpenAI announced GPT-4 API general availability and deprecated older completion models (GPT-3 base models and text-davinci-003), requiring developers to migrate to new models like gpt-3.5-turbo-instruct or newer by January 4, 2024. Fine-tuned models will need to be retrained on new base models with priority access offered for GPT-3.5 Turbo and GPT-4 fine-tuning.
OpenAI announced its first Asian office in Tokyo, Japan, and released a GPT-4 custom model optimized for Japanese language with 3x faster performance and 47% reduced token costs. The expansion includes appointing Tadao Nagasaki as President of OpenAI Japan and partnerships with major Japanese companies and government institutions.
Zelma, a GPT-4 powered research assistant developed by Dr. Emily Oster and her team at Brown University in partnership with Novy, makes standardized test data accessible to parents, teachers, administrators, and policymakers by allowing natural language queries about student performance across districts and demographics.
Lifespan health system used GPT-4 to simplify surgical consent forms from three pages to one page at a 6th grade reading level, improving patient understanding and physician adoption. The initiative, deployed in September 2023, has received positive feedback from both patients and clinicians.
Paradigm leverages GPT-4's natural language understanding to dramatically improve patient screening for clinical trials, enabling evaluation of hundreds of patients per minute compared to manual review of ~50 per day, reducing clinician burden and improving patient access to treatments.
WHOOP launched WHOOP Coach, an LLM-powered health coaching assistant powered by GPT-4, delivering personalized fitness and wellness recommendations to wearable users based on their individual data. The product marks a significant application of large language models in the health and wellness sector, with strong member engagement around sleep improvement and self-optimization.
Summer Health uses OpenAI's GPT-4 to automatically convert pediatric visit notes from medical shorthand into clear, jargon-free summaries for parents, reducing administrative burden on doctors while improving accessibility and reducing wait times for caregivers.
Ironclad integrated OpenAI's GPT-4 into its AI Assist™ contract review tool, enabling legal teams to automate contract editing while maintaining human oversight and data privacy. The integration has been live since GPT-4's launch in April 2023.
Typeform launches Formless, an AI-powered platform built on OpenAI's GPT-3.5 Turbo and GPT-4 that replaces traditional forms with dynamic conversational interactions. The platform enables natural, two-way dialogue for data collection where respondents can ask follow-up questions like in human conversation.
OpenAI publishes an explainer on its core technology, detailing how language models like GPT-4 are developed through pre-training (learning from vast text data) and post-training (alignment with human values and safety practices). The article emphasizes OpenAI's nonprofit mission structure and explains the distinction between raw base models and refined, usable versions.
OpenAI is launching ChatGPT Enterprise, offering enterprise-grade security and privacy, unlimited GPT-4 access, 32k context windows, advanced data analysis, and admin controls, with a guarantee that business data is not used for model training.
OpenAI partners with Scale to provide enterprise support for fine-tuning models, enabling companies to securely customize GPT-3.5 and GPT-4 on proprietary data with Scale's data enrichment and evaluation expertise.