“To be clear, we don’t train our models on fake or hallucination data,” says Waseem Alshikh, co-founder of Writer. This dedication to authenticity has propelled Writer from its initial $21 million seed funding in 2021 to an impressive valuation of $1.9 billion as the company seeks to raise up to $200 million in its latest funding round.
Writer has made headlines not just for its valuation, but also for its unique approach to AI model training. The company reported that it spent approximately $700,000 on developing its latest model, a stark contrast to the millions typically invested by competitors like OpenAI and Anthropic. This cost-effective strategy involves the use of synthetic data—AI-generated information designed to replicate real-world data while protecting privacy.
Waseem Alshikh, Writer’s co-founder and CTO, emphasized the clarity of their synthetic data approach. “There’s some confusion in the industry about the definition of ‘synthetic’ data,” he noted. “To be clear, we don’t train our models on fake or hallucination data, and we don’t use a model to generate random data. We take real, factual data and convert it to synthetic data that is specifically structured in a clearer and cleaner way for model training.”
Industry Context and Competitive Landscape
Writer’s approach comes at a critical juncture in the AI landscape. A recent study by AI researchers indicates that the available public training data could be exhausted between 2026 and 2032, highlighting the increasing reliance on synthetic data in AI development. Notably, major companies like Amazon, Meta, and Microsoft-backed OpenAI have begun integrating synthetic data into their training processes.
However, experts caution against the potential pitfalls of synthetic data, which could degrade model performance or exacerbate existing biases if mismanaged.
Palmyra X 004
Palmyra X 004 stands out in the crowded generative AI space for its impressive performance in function calling and workflow execution. It achieved a score of 78.76% on the Berkeley Tool Calling Leaderboard, outperforming models from tech giants by nearly 20%. This model’s capabilities extend beyond mere text generation; it allows enterprises to create AI-enabled applications capable of performing complex tasks, analyzing data, and executing workflows autonomously.
“We’re enabling AI to execute multiple functions and actions simultaneously, which is crucial for automating complex enterprise workflows,” Alshikh stated. “With Palmyra X 004, we’re moving from AI assistants that simply provide information to systems that can actually do work.”
The model supports multilingual capabilities across 30+ languages and boasts a 128,000 token context window, allowing it to process lengthy documents and conversations effectively. Additionally, it can handle multimodal inputs, including text, images, and audio, although the latter capabilities are still in beta.
Clientele and Market Potential
Writer serves over 250 enterprise clients, including heavyweights like Accenture, Uber, Salesforce, L’Oréal, and Vanguard along with founder May Habib. The generative AI market is projected to surpass $1 trillion in revenue within the next decade, with investors pouring $26.8 billion into 498 generative AI deals in 2024 alone. Companies in this sector raised $25.9 billion in 2023—an increase of over 200% from the previous year.
Alshikh pointed out the broader implications of Writer’s efficiency-focused approach, suggesting it could pave the way for more affordable and accessible enterprise AI solutions. “We’ve found a way to build highly capable models without relying on massive parameter counts or exorbitant training costs. Our model training costs were below a million dollars in GPU time for something above 100 billion parameters. We’re proving that you don’t need hundreds of billions of dollars to compete in the AI race.”
Future Directions and Challenges
While Writer’s advancements are notable, the company acknowledges ongoing challenges in the AI landscape. As systems become more integrated into business processes, issues of reliability, explainability, and governance grow increasingly important. Writer has sought to address these concerns by incorporating features like automatic data integration with retrieval augmented generation (RAG) and source transparency.
The company emphasizes the importance of AI safety and control, with Palmyra X 004 integrating with Writer’s existing suite of AI guardrails and governance tools, allowing enterprises to set content policies and manage the model’s outputs.
Looking forward, Alshikh hinted at future research directions, including the exploration of deeper transformer models that could enhance reasoning capabilities. “We’re at an inflection point in AI development,” he stated. “The next frontier isn’t just about making models bigger, but making them smarter and more efficient. We’re focusing on architectural innovations that can deliver better reasoning at lower inference costs.”