Software companies are constantly trying to add more and more AI features to their platforms, and it can be hard to keep up with it all. We’ve written this roundup to share updates from 10 notable companies that have recently enhanced their products with AI.
Google announces public beta for Gen AI Toolbox for Databases
Gen AI Toolbox for Databases is an open-source server that lets developers connect generative AI applications to databases.
According to Google, the toolbox can help streamline creation, deployment, and management of generative AI tools that are capable of querying databases.
Currently the toolbox provides connectivity to self-managed PostgreSQL and MySQL as well as managed databases AlloyDB, Spanner, Cloud SQL for Postgres, Cloud SQL for MySQL, and Cloud SQL for SQL Server.
JetBrains AI Assistant can now use local LLMs
JetBrains has announced an update to its AI Assistant that enables it to use local models so that developers under strict data privacy and compliance rules can work with it.
To enable this feature, users will need to enable LM Studio under “Third-party AI providers” in the AI Assistant settings. LM Studio provides an interface for managing and running AI models on a local machine.
This latest release also provides access to the latest models from Anthropic and OpenAI, including Claude 3.5 Sonnet and Claude 3.5 Haiku, and OpenAI’s o1, o1-mini, and o3-mini.
Gemini Advanced users can now reference past chats
By recalling past conversations, Gemini will be able to provide more helpful responses, Google claims. It will also allow users to build on previous conversations or projects.
Users can control what information is stored, and will be able to review, delete, and decide how long to keep chat history.
This feature is currently only available in English for Gemini Advanced subscribers that are part of the Google One AI Premium Plan, but Google says it will roll out to additional languages and Google Workspace Business and Enterprise users in a matter of weeks.
Gemini API adds access to Imagen 3
Imagen is Google’s image generation model, and version 3 can now be accessed by paid users through the Gemini API, with a rollout to the free tier to follow.
According to Google, Imagen 3 achieved “state-of-the-art performance” across a number of benchmarks and features improved prompt following.
Additionally, all images created using Imagen 3 will contain a non-visible digital SynthID watermark that identifies them as having been generated by AI.
Adobe announces generative AI video model
Adobe Firefly Video Model allows content creators to generate videos using either image or text prompts.
The model was trained on licensed content, including Adobe Stock and public domain content with expired copyright, enabling creators to use Firefly output without worrying about copyright or IP issues.
“We are working to create a future where creators’ rights are protected, there is transparency around how content is made and edited, and customers have the choice to use our Firefly models that are IP-friendly and commercially safe as well as the option to leverage third-party models safely in our tools and services,” Adobe wrote in a blog post.
Workday announces centralized system for managing AI agents
Workday Agent System of Record manages Workday and third-party agents in one place, offering the ability for customers to onboard new agents, define roles and responsibilities, turn them on or off, track impact, manage costs, and ensure compliance.
“Agent System of Record gives IT and business leaders the foundation of trust, transparency, and control they need to understand the true value of their AI investments, while keeping their organizations resilient and secure in this new era of human-machine collaboration,” Workday wrote in a blog post.
Snowflake announces public preview of Cortex Agents
Cortex Agents are AI agents that work on enterprise data sources, both in structured and unstructured formats. According to the company, they can help break down complex queries, retrieve relevant data, and generate answers, utilizing the help of Cortex Search, Cortex Analyst, and LLMs.
“Cortex Agents, using Cortex Analyst, Cortex Search, Anthropic’s Claude models and AI Observability, bring intelligence on top of a unified governance framework and efficient processing engine for both structured and unstructured data. Using these building blocks, developers can build and deploy data agents that can be integrated to their application of choice using the REST API interface,” Snowflake wrote in a blog post.
Mabl launches mabl GenAI Test Creation
Mabl GenAI Test Creation uses generative AI to automatically create tests based on existing assets. The tests it generates are designed to be easy to maintain in order to cut down on technical debt in the testing process.
“mabl’s GenAI Test Creation is changing how we approach and execute work,” said Randall Morse, senior principal software engineer at Barracuda Networks. “It allows us to generate test cases from just a few words and automatically create a test plan with assertions, accelerating our development and testing cycle. Its ease of use helps our junior engineers extend their reach while freeing up senior developers to tackle more complex challenges.”
Ai2 announces iOS app for its OLMoE model
OLMoE is an open source toolkit that developers can use to experiment with on-device AI, such as experiencing what tasks on-device AI can handle, researching how to improve local AI models, or testing a model locally using Ai2’s open source codebase.
According to Ai2, developers can also integrate OLMoE into other iOS applications. OLMoE models are fully private, meaning prompts and responses do not leave the device. They also do not require an internet connection, so they can be used regardless of a user’s connectivity.
“We see this as a foundational step towards the future of on-device functionality. As mobile devices continue to increase in power and performance, we hope that the OLMoE app can help researchers and developers keep up with the cutting edge,” Ai2 wrote in a blog post.
DataRobot acquires Agnostiq
DataRobot believes this acquisition — which includes Agnostiq’s open-source infrastructure management and compute orchestration platform Covalent — will enable it to further advance agentic AI development.
Key benefits that Agnostiq brings include heterogeneous compute orchestration, reduced AI infrastructure complexity and costs, and an open and agnostic approach to tools and infrastructure.
“Agentic AI applications introduce new levels of complexity—from managing multi-agent workflows to orchestrating diverse compute environments. With Covalent now part of DataRobot, AI teams can focus on building, not infrastructure,” DataRobot wrote in a blog post.