More On: Google Natural Language
Google Cloud offers developers access to its foundational models
Global Market Analysis and Forecast for Data Science and Machine Learning Platforms: Key Players, Regional Trends, and Future Outlook until 2029
Essential Machine Learning Knowledge for Data Scientists
Forecasting Machine Learning as a Service Market: Size, Segmentation, Parameters, and Projections through 2032
Deep Learning Innovations Fuel the Expansion of the Machine Learning Industry - Grand View Research, Inc.
Google is entering into competition with other AI players as it launches a significant language model API and additional developer tools.
On Tuesday, Google Cloud made a significant move by making some of its AI-powered applications available to developers. The highlight of this release is the Pathways Language Model, or PaLM, API, which is specifically designed for language models. The release also includes the Makersuite prototyping tool, which is integrated into the API. These developer tools work in conjunction with Google's new AI productivity tools, which are currently in the testing phase in Google Workspace. With this technology, the productivity AI can create business documents such as emails and job descriptions upon request.
For developers who are eager to work with these tools, Google has introduced new ways to access them. However, the PaLM API's efficient model, which is highly capable in terms of size and performance, is currently only available to select developers as part of the Private Preview program. Nonetheless, Google is expected to open a waitlist soon for other interested developers to join.
Google opens gateway to PaLM API
PaLM is a cutting-edge language model that excels at understanding natural language prompts and generating coherent responses. The PaLM API provides developers with the tools they need to build applications that leverage this technology. Specifically, the API includes models that are optimized for multiturn use cases like content generation and chat, as well as more general-purpose models designed for summarization, classification, and other tasks.
In a recent blog post, Google announced that an "efficient model...in terms of size and capabilities" is currently available, with larger models expected to be released in the near future. However, the efficient model is currently only accessible to a select group of developers who are part of the Private Preview program. Other interested developers will need to join a waitlist in order to gain access to this technology.
What is Google’s Makersuite?
Google has introduced Makersuite, a low-code tool that is part of the API, to help developers quickly create prototypes and experiment with potential use cases. Google has also announced plans to include additional features for prompt engineering, custom-model tuning, and synthetic data generation to streamline generative AI development workflows and reduce the need for developers to switch between different tools.
Makersuite is not only useful for developers but also allows business owners to tailor the API to their specific needs. Once completed, developers can export their prompts in widely used languages and frameworks such as Python and Node.js, facilitating integration into their existing workflows.
In a blog post, Google's Scott Huffman, Vice President of Engineering, and Josh Woodward, Senior Director of Product Management, emphasized their commitment to providing a unified development environment. They acknowledged that generative AI development workflows can be fragmented, with developers having to use multiple tools to accomplish tasks like crafting and iterating on a prompt, generating synthetic data, and tuning a custom model.
Moreover, Google has prioritized the safety of its AI systems and ensured that the API and tool comply with its AI safety standards. The company has also promised that training data generated by non-Google developers using the API will remain confidential and will not be fed back into the large language model, providing reassurance to users concerned about data privacy.
Other AI tools for Google Cloud enterprise customers
Google is attempting to challenge Microsoft and OpenAI's dominance in the natural language AI chat space. On Tuesday, the company announced several new AI features for its enterprise cloud customers.
One of these features is the Generative AI App Builder, which allows users to create apps that utilize conversational AI for a variety of purposes. This tool connects directly to Google's out-of-the-box search capabilities and foundation models, providing enterprises with a powerful solution for developing conversational AI applications.
Google Cloud's Vertex AI platform, which allows enterprises to build and deploy machine learning models and AI applications at scale, has also received a significant upgrade. Vertex AI now has access to foundation models, enabling enterprise customers to discover models, create and modify prompts, and fine-tune those prompts using their own data. Currently, Vertex AI can create text and images, but Google expects to add support for video and audio in the future.
These new AI features demonstrate Google's commitment to providing its enterprise customers with powerful tools for developing AI applications. By integrating natural language processing capabilities into its cloud offerings, Google hopes to challenge the dominance of its competitors in the natural language AI chat space.
Source Techrepublic
