Tech companies like OpenAI are being accused of spreading false narratives about the dangers and costs of building generative AI platforms, in an attempt to maintain their dominance in the field. Databricks’ CTO, Matei Zaharia, believes that these companies are obstructing the democratization of artificial intelligence (AI) by discouraging other companies from participating in AI development. Zaharia argues that the narrative being pushed by firms like OpenAI, which highlights the potential risks and high costs associated with generative AI, is not necessarily true.
Zaharia points to MosaicML, a startup recently acquired by Databricks, as an example of a company that has successfully trained a large language model (LLM) with 30 million parameters. This model competes with OpenAI’s GPT-3 but at a significantly lower cost. Zaharia believes that the pace of change and innovation in the AI industry challenges the idea that only a few players can understand and control the technology to make it safe. He highlights the speed at which other companies have replicated and even surpassed OpenAI’s models at a lower cost.
Zaharia suggests that as more enterprises experiment and discover new use cases for LLMs and other generative AI systems, the technology will become more widespread and general. He believes that the fear-mongering about the dangers and costs of generative AI is a tactic employed by tech giants to maintain their stranglehold on the emerging field.
The narrative being promoted by OpenAI and other tech firms implies that generative AI is a disruptive and potentially evil technology that requires extensive work and investment to train models. However, Zaharia challenges these claims, arguing that companies like MosaicML have demonstrated that it is possible to achieve comparable results at a fraction of the cost.
The AI industry is experiencing a rapid pace of change, similar to previous technological advancements like blockchain and cloud computing. Many companies are eager to enter the AI space, whether it is to implement AI technology or develop their own generative AI systems. Zaharia encourages these companies to question the false narratives being spread and to explore the possibilities offered by generative AI.
In conclusion, tech giants like OpenAI are being accused of misleading the industry by exaggerating the dangers and costs of generative AI, thereby discouraging competition and hindering the democratization of AI. However, smaller companies are proving that it is possible to achieve comparable results at a lower cost. As the technology continues to evolve, more enterprises are expected to experiment and discover new use cases for generative AI systems. The article emphasizes the need for companies to question false narratives and seize the opportunities presented by generative AI.