Harnessing the power of Generative AI to protect your data

Harnessing the power of Generative AI to protect your data

by | Sep 5, 2024 | Technology

This article is contributed. See the original author and article here.

In today’s digital era, where data breaches and cyber threats are increasingly sophisticated and pervasive, the need for robust data security measures has never been more critical. Traditional security approaches are proving insufficient against the complex and evolving nature of modern cyber threats. This has led to a growing consensus among security experts and industry leaders on the imperative to incorporate Generative AI (GenAI) into data security frameworks. GenAI’s ability to analyze vast amounts of data in real-time, identify patterns, and predict potential threats offers a transformative approach to safeguarding sensitive information. According to a recent report by Gartner, the use of AI in cybersecurity is expected to reduce the number of data breaches by up to 20% by 2025, underscoring the industry’s recognition of AI’s vital role in enhancing data security (Gartner, 2022). This blog explores how Microsoft is leveraging GenAI to revolutionize data security, providing organizations with the tools they need to protect their digital assets effectively.


 


Leverage the power of Copilot to secure your organization


Human ingenuity and expertise will always be an irreplaceable component of defense, so we need technology that can augment these unique capabilities with skill sets, processing speeds, and rapid learning of AI. Technology that can work alongside us, detect hidden patterns and behaviors, and inform response at machine speed with the latest and most advanced security practices.


 


In this scenario, Microsoft Copilot for Security helps professionals across the many cybersecurity disciplines to be more effective and efficient at all the roles they play. It helps you enhance and grow your capabilities and skills, while also supporting the workflows and teams you collaborate with to solve security challenges. Since Copilot for Security uses GenAI to analyze data from many sources, including other Microsoft Security solutions, it can also help analysts catch what they might have otherwise missed. Copilot for Security synthesizes data and detects those important signals better than ever before, all in a single pane of glass, without having to jump between different solutions to get additional context.


 


Boost your data protection efficiency with Copilot for Security embedded in Purview


An important application of Copilot for Security is to empower and strengthen data security and data compliance teams in securing data with more efficiency and agility. Data security admins are often challenged by the high volume and complexity of alerts, and the integration between Microsoft Purview and Copilot for Security enables these tools to work together to protect your data at machine speed.


 


The speed at which data security investigations are conducted is crucial to preventing data loss. However, the task of analyzing a vast array of sources can pose a significant challenge for analysts at any experience level. With Copilot-powered comprehensive summaries of Microsoft Purview Data Loss Prevention (DLP) alerts, data security admins can identify, act on alerts and prevent data risks much faster and effectively. When an alert is summarized, it includes details such as policy rules, the source, and the files involved, as well as user risk level context pulled from Insider Risk Management (IRM).


 


Figure 1: embedded Copilot summarization into Data Loss PreventionFigure 1: embedded Copilot summarization into Data Loss Prevention


Your team can also leverage summaries in Microsoft Purview Insider Risk Management alerts, which enables faster understanding of potentially risky activity, user context, behavior sequences. and intent present in an alert. Moreover, we’re excited to announce the public preview of the Copilot for Security-powered enhanced hunting in IRM, where admins will be able to use GenAI-driven analytics to deepen investigations and double-click into a user’s risk profile and activities, beyond the alert summary.


 


Figure 2: embedded Copilot summarization into Insider Risk ManagementFigure 2: embedded Copilot summarization into Insider Risk Management


Compliance admins, forensic investigators, legal, and other teams can also strongly benefit from GenAI being incorporated into their workflows. Not only do they spend most of their time reviewing lengthy content and evidence; but admins need to invest time to learn complex technical capabilities like keyword query language to conduct a search, with 60% of admin time spent reviewing evidence collected in review sets.


 


Compliance teams are subject to regulatory obligations, like industry regulations or corporate policies related to business communications. This requires teams to review communication violations that contain lengthy content like meeting transcripts, group chats, long email threads and attachments. With concise and comprehensive contextual summaries on Microsoft Purview Communication Compliance, content can be evaluated against relevant compliance polices and investigators are able to get a summary of the policy match and better identify risky communication.


 


Figure 3: embedded Copilot summarization into Communication ComplianceFigure 3: embedded Copilot summarization into Communication Compliance


These contextualized summaries are also invaluable in Microsoft Purview eDiscovery, where they help simplify the exploration of large about of evidence data, which can take hours, days, even weeks to do. This process often requires costly resources like an outside council to manually go through each document to determine relevancy to the case, and this embedded Copilot for Security capability enables reducing days of legal investigations into seconds, by allowing an investigator to use Copilot to summarize items in a review set.


 


Figure 4: embedded Copilot summarization into eDiscoveryFigure 4: embedded Copilot summarization into eDiscovery


Search is one of the most difficult and time-intensive workflows in an eDiscovery investigation. Now, you can simplify investigation by leveraging Copilot for Security to translate inquiries from natural to keywork query language. This feature allows organizations to take Natural Language and convert that into assertive evidence queries, in doing so this can correct possible errors, boost team expertise, and enable analysts at all levels to carry out advanced investigations.


 


Figure 5: embedded Copilot search with Natural Language on eDiscoveryFigure 5: embedded Copilot search with Natural Language on eDiscovery


All these recent developments are just the beginning of the Copilot for Security journey into Microsoft Purview, and we’ll continue to share new advancements and GenAI-powered capabilities that will take your data security program to the next level.


 


To learn more about how Microsoft Purview can help you protect your data, check our website, or explore our learning resources focused on Copilot for Security in Purview.


 


Get Started


What’s new in Forms for EDU – Sep 2024

What’s new in Forms for EDU – Sep 2024

by | Sep 4, 2024 | Technology

This article is contributed. See the original author and article here.

With the new semester just around the corner, we’re excited to introduce some fantastic new features in Forms to enhance your teaching and learning experience. Now, you can enjoy an improved data sync function to stay up to date, set rules to collect more accurate responses, pre-fill forms and quizzes, and easily manage response notifications. Let’s dive in and check more details. You can also try out from this template.


 


Let’s create a science club registration together to explore how we can make the most of these new features. Imagine you are a high school science teacher and trying to establish a science club where students from all grades can dive into fascinating topics beyond the curriculum and participate in fun, hands-on activities.


 


Improved Excel data sync function


You can start with Excel for the web to create the registration form, no need to visit Forms. This process will automatically generate a new form in Forms where you can add your questions. All the questions you add and responses you receive will seamlessly sync with Excel. (check more details: Forms data sync to Excel is now fully available with more functionality – Microsoft Community Hub)


 


Create forms from Excel for the webCreate forms from Excel for the web


Prefill answers for forms/quizzes


In this science club registration, you can use prefilled answers for fields like grade and class, creating unique links for each grade and class. This allows you to easily track and differentiate student distribution across different groups in the club. (check more details: Pre-fill Responses in Your Microsoft Forms – Microsoft Community Hub)


 


Set prefilled answerSet prefilled answer


Validation for open text question


To make sure students provide correct contact information and receive timely updates, you can set restriction for open text questions to only accept email formats. (Other types of restriction are also supported in Forms, check more details here: Validate open text responses in Microsoft Forms – Microsoft Community Hub)


 


Set validation on email addressSet validation on email address


Manage recipients of new response notification


You can set multiple key individuals to receive the notification when a new registration comes in, such as the club coordinator and an assistant. This allows them to quickly review and confirm new members, update the membership list, and prepare welcome materials. (check more details: Add or remove people from your form response notification emails – Microsoft Community Hub)


 


Add recipients for new response notificationAdd recipients for new response notification


Hope you find these new features useful for the new semester. If you’re interested in trying them out, simply start with this template.

All in on AI: Exploring Microsoft’s AI journey through customer service

All in on AI: Exploring Microsoft’s AI journey through customer service

by | Sep 4, 2024 | Dynamics 365, Microsoft 365, Technology

This article is contributed. See the original author and article here.

All in on AI is a series featuring interviews from Microsoft executives across the company about what transforming work with AI means to their teams. Through these conversations, we’ll highlight the challenges each industry faces and the lessons we learned that our customers can use in their own AI journeys. In this episode, Colette Stallbaumer, General Manager, Microsoft 365 and Future of Work sits down with Mala Anand, Corporate Vice President, Customer Experience and Success at Microsoft to explore the integration of AI in customer service.

According to the 2024 Work Trend Index, a study exploring into the global trends affecting employee work and wellness, employees are struggling to keep up with the pace and volume of work. Many employees report feeling burned out and as a result, staff turnover is on the rise.

These trends have profound implications, underscoring a crucial shift: employees are now prioritizing their health and well-being over work. This is a pressing concern for all business leaders, particularly in customer service. 

At Microsoft, we recognize the vital role our customer service agents play in delivering world-class care. As the front line between our business and our customers, the quality of our customer care directly impacts customer satisfaction. 

Recently, Colette Stallbaumer, General Manager, Microsoft 365 and Future of Work joined Mala Anand, Corporate Vice President, Customer Experience and Success at Microsoft to discuss how our customer service teams are using AI solutions like Microsoft Copilot to ensure our customer service employees are empowered to focus on the most meaningful parts of their jobs. Here’s what she had to say: 

Watch the full interview

Adapting to the customer service surge with the power of AI 

Since 2019, Microsoft’s customer service teams, like many in the industry, have been managing a rising surge in calls from customers with complex needs. 

As a service industry veteran and leader of our customer service organization, Mala and her team realized that just adding more call-handling capacity wouldn’t suffice. Their goal was to help teams at scale—to empower them to manage more inquiries, address more complex customer needs, and to more rapidly onboard new employees to help with the surge. They also saw an opportunity to reduce case volumes by introducing self-service options for customers. 

AI strategy roadmap


Get the e-book

The team began exploring the use of generative AI to enable a more consistent and seamless experience. Already one of the world’s largest support organizations using Dynamics 365 Customer Service, harnessing new AI capabilities could help the organization to automate and augment common support scenarios. As customer zero for our products, turning to Microsoft Copilot and Dynamics 365 Customer Service would enable us to quickly adopt the transformative power of generative AI, and share those learnings with business leaders to implement and improve their own customer support experience at scale.

Reimagining the customer experience with the power of AI

Stay competitive in today’s market with AI


Read more from Mala

Transforming the employee experience with generative AI 

As Mala explains in the interview, adopting Copilot in Dynamics 365 Customer Service and Copilot Studio created an opportunity to transform work across four key areas of Microsoft’s customer service landscape. 

First, the Customer Service and Success team needed to speed up how quickly new hires could start helping existing customer service agents. Second, they wanted all support teams to access accurate information quickly and spend less time working across tools and data sources to collect customer and incident details. Third, they aimed to cut down on mundane tasks like data entry and paperwork so that customer service agents could focus more on complex issues that require empathy and judgment. Lastly, they aimed to improve self-help options for customers to handle their own questions, reducing the need for support requests. 

AI has the potential to drive operational excellence, nurture customer loyalty, and grow value through the entire customer journey—making it one of the most valuable tools for enterprises to create a consistent, differentiated experience, build meaningful relationships, and deliver better customer care.

Mala Anand, Corporate Vice President, Customer Experience and Success at Microsoft

1. Supporting new hires with better onboarding

The Microsoft Customer Service team introduced Copilot’s case summarization feature to break down complex information into simple summaries, helping new hires understand customer issues more quickly. By surfacing relevant information from diverse sources such as internal documentation, knowledge bases, and websites, Copilot accelerates the learning curve, enabling agents to resolve customer queries faster and deliver superior service. 

What’s more, Copilot supports new hires with ongoing learning and skill development by providing real-time assistance and insights. By using AI capabilities integrated into platforms like Dynamics 365 Customer Service, agents can more easily access tools and diagnostics and recommend optimal solutions. This empowers agents to deliver highly personalized customer experiences across all customer support needs, at a global scale.

Explore Copilot’s impact on onboarding efficiency

2. Helping employees find the right information faster

With Copilot, agents can now find important information quickly, even if it was spread across different systems. Dynamics 365 Knowledge Assist helps agents zero in on the relevant information and articles faster than traditional search. It not only can suggest relevant information but provides context about the specific relevance of content and cites the source.

Having a more comprehensive and streamlined method to seek information not only helps agents solve issues more quickly—it gives them a more complete understanding of information relevant to the issue they’re trying to solve. Agents can now use this context, with Copilot, to craft a more thorough and accurate response to customers.

Enhance workflow efficiency with Copilot-powered search

3. Automating burdensome tasks

Before Copilot, creating detailed case summaries could take up to 30 to 40 minutes. Now, with Copilot, it’s done in 15 minutes or less. This not only speeds up how quickly customer service agents can handle customer requests but also lets them manage more cases without sacrificing quality. 

Automating summaries and common email and communication tasks allows agents to spend much less time on repetitive tasks, giving them more time to focus on the more meaningful and often more rewarding aspects of their job.  

Discover effortless task automation with Copilot

4. Transforming support with self-service

Prioritizing self-service by integrating Copilot Studio into our customer-facing experiences is starting to make a big difference in how Microsoft handles customer service. Now, customers can use the power of generative AI in Copilot Studio to access self-help resources.

This change has reduced the amount of work customer service teams have to do and lets agents concentrate on more complex issues where customers require assistance. As a result, they’re getting things done more efficiently, working together better, and feeling more satisfied with their jobs. 

Learn to empower customers with self-service solutions

Harnessing the business impact of Copilot  

As Mala explains in her interview, generative AI and Copilot have played a pivotal role in transforming our customer service organization at Microsoft, enabling teams to achieve more with fewer resources. Here’s what we’ve seen so far:

  1. Improved collaboration: Collaboration required among customer support agents decreased by 13%, making case resolution smoother and operations more efficient.
  1. Enhancing efficiency: With Copilot, average chat case handling time reduced by up to 16%, giving agents better access to knowledge and improving overall efficiency.
  1. Increasing productivity: Agents using Copilot became more productive, handling up to 12% more cases, which allowed them to serve more customers effectively. 

And that’s just the start. Mala sees a future for Copilot that will revolutionize the customer services industry, making it more efficient, responsive, and customer-centric.  

These include things like automatically connecting customers with appropriate experts, helping customers solve complex problems on their own, predicting issues before they happen, and personalizing interactions based on customer information and preferences. 

Copilot learning hub

Learn how to use Microsoft Copilot based on your role


Get started

Join Microsoft on our AI journey  

By automating routine tasks and helping teams find information faster, Copilot helps our employees focus on what really matters to them, allowing them to dive deeper into the parts of their jobs they love.  

The speed at which AI is transforming business today is unprecedented, reimagining the way we work. We envision a future where AI alongside human expertise reduces and eliminates the seams in the support experience. 

Mala Anand, Corporate Vice President, Customer Experience and Success at Microsoft

Today, the power of AI is revolutionizing how companies innovate and create large-scale impact in all areas. Leveraging generative AI throughout the customer journey enables businesses to foster enduring customer loyalty by improving the overall experience, adding value at each interaction. 

Discover how AI is transforming customer service at Microsoft

This interview is part of our All in on AI series that explores how Microsoft is adopting AI across different areas of our business. Next, Kathleen Hogan, Microsoft Chief People Officer, sits down with Jared Spataro to share how AI is helping human resource teams do more with less and to share the transformative best practices she used to drive some of the highest, fastest AI adoption rates Microsoft has ever seen. 

The post All in on AI: Exploring Microsoft’s AI journey through customer service appeared first on Microsoft Dynamics 365 Blog.

Brought to you by Dr. Ware, Microsoft Office 365 Silver Partner, Charleston SC.

Try the first-party WhatsApp channel in Dynamics 365 Contact Center

Try the first-party WhatsApp channel in Dynamics 365 Contact Center

by | Sep 3, 2024 | Dynamics 365, Microsoft 365, Technology

This article is contributed. See the original author and article here.

In today’s digital landscape, the success of customer service on social media hinges on the quality of care provided to customers. It’s essential for communications from agents to be timely, accurate, sensitive, brief, and friendly to enhance customer satisfaction and foster brand loyalty. To support this, we are excited to introduce the first-party WhatsApp channel seamlessly integrated with Azure Communication Services, now available in public preview for Dynamics Contact Center. 

Why WhatsApp?

WhatsApp is a leading choice for contact centers worldwide, celebrated for its ease of use and extensive global reach. With billions of users, WhatsApp provides businesses with an unparalleled opportunity to connect with customers in a medium they trust and prefer. Now, with our first-party solution, integrating WhatsApp into your contact center has never been easier. 

Seamless integration with Microsoft’s ecosystem 

One of the key advantages of this new integration is its direct tie-in with Microsoft’s ecosystem. By leveraging Azure Communication Services, our solution eliminates the need for third-party connectors. This results in a simplified IT footprint, enhanced reliability, and more secure communication for your business. No more juggling multiple vendors or worrying about potential integration issues. Our solution is designed to work seamlessly within your existing Microsoft infrastructure. 

Key capabilities 

With the first-party WhatsApp channel, your contact center can benefit from a range of powerful features: 

  • Easy Configuration: Administrators can easily configure the WhatsApp channel powered by Azure Communication Services, making setup straightforward. 
  • Customer Engagement: Customers can reach out to your business through WhatsApp, engaging in an asynchronous manner that respects their time and preferences. Whether it’s a quick question or a detailed inquiry, your customers can connect with you when it’s convenient for them. 
  • Unified Interface for Agents: Agents will have access to a unified, contextual, and productive interface. Subsequently, they can engage with customers and resolve issues more efficiently. This streamlined workflow reduces the time to resolution and enhances overall customer satisfaction. 
  • Rich Reporting for Supervisors: Supervisors and managers can access detailed reports to monitor and optimize contact center performance. This data-driven approach ensures that your team is operating at peak efficiency, with insights that drive continuous improvement. 

Enhancing customer engagement

The integration of WhatsApp with Azure Communication Services is more than just a new channel—it’s a step toward more meaningful and contextual customer interactions. By leveraging WhatsApp, businesses can provide a seamless and personalized experience that resonates with customers, ultimately driving satisfaction and loyalty.

Get started today

To sign up to use this feature, fill out this form. We invite you to explore the new first-party WhatsApp integration in Dynamics 365 Contact Center. Whether you’re addressing product inquiries or managing customer service scenarios, this integration empowers your agents to deliver exceptional customer care. 

Join us in this exciting journey of enhancing customer communication. With this new channel, experience the benefits of a simplified, reliable, and powerful customer service tool. 

Stay tuned for more updates as we continue to expand the capabilities of Dynamics 365 Contact Center. Also, don’t hesitate to provide feedback during this public preview period. We’re here to support your success. 

Learn more

Watch a quick video introduction.

To learn more, read the documentation: Configure a WhatsApp channel through Azure Communication Services (preview) | Microsoft Learn

The post Try the first-party WhatsApp channel in Dynamics 365 Contact Center appeared first on Microsoft Dynamics 365 Blog.

Brought to you by Dr. Ware, Microsoft Office 365 Silver Partner, Charleston SC.

Explore Azure AI Services: Curated list of prebuilt models and demos

by | Sep 3, 2024 | Technology

This article is contributed. See the original author and article here.

Explore Azure AI Services: Prebuilt Models and Demos


Azure AI services provide a comprehensive suite of prebuilt models and demos designed to address a wide range of use cases. These models are readily accessible and allow you to implement AI-powered solutions seamlessly. We have curated and catalogued prebuilt demos available across Azure AI services. We hope this helps you infuse AI seamlessly into your products and services.


 


Speech Recognition


 


Speech to Text Scenarios


 










































Scenario Description Link
Real-time speech to text Quickly test your audio on a speech recognition endpoint without writing any code. Explore Demo
Whisper Model in Azure OpenAI Service Transcribe and translate audio content from 57 languages into English using OpenAI Whisper v2-large model. Explore Demo
Batch speech to text Transcribe large amounts of audio in storage asynchronously. Explore Demo
Custom Speech Improve speech recognition accuracy with domain-specific vocabulary and data. Explore Demo
Pronunciation Assessment Evaluate and get feedback on speech pronunciation accuracy and fluency. Explore Demo
Speech Translation Translate speech into other languages in real-time with low latency. Explore Demo

 


Text to Speech


 


Text to Speech Scenarios


 





































Scenario Description Link
Voice Gallery Choose from 486 voices across 148 languages and variants to create natural-sounding speech. Explore Demo
Custom Neural Voice Create a natural-sounding synthetic voice based on human voice recordings. Explore Demo
Personal Voice Create an AI voice from a human voice sample for personalized voice experiences. Explore Demo
Audio Content Creation Build highly natural audio content for various scenarios like audiobooks and video narrations. Explore Demo
Text to speech Avatar Turn your text into a video with an AI-generated avatar and realistic voice. Explore Demo

 



Other Scenarios




 







































Scenario


Description


Link



Captioning with Speech to Text


Use our sample application to learn how to use Azure Speech to automatically caption your content in real-time and offline by transcribing the audio of films, videos, live events, and more. Display the resulting text on a screen to provide an accessible experience. In this example, we leverage features like speech to text and phrase list.




Post Call Transcription & Analytics


Batch transcribe call center recordings and extract valuable information such as Personal Identifiable Information (PII), sentiment, and call summary. This demonstrates how to use the Speech and Language services to analyze call center conversations.




Live Chat Avatar


Engage in natural conversations with an avatar that recognizes users’ speech input and responds fluently with realistic AI voice.




Language Learning


Get instant feedback on pronunciation accuracy, fluency, prosody, grammar, and vocabulary from your chatting experience.




Video Translation


Seamlessly translate and generate videos in multiple languages automatically. With its powerful capabilities, you can efficiently localize your video content to cater to diverse audiences around the globe.









 


Vision Studio


 


Vision-Based Scenarios


 








































































Scenario Description Link
Video Retrieval and Summary Quickly summarize the main points of a video and search for specific moments. Explore Demo
Customize Models with Images Find specific objects within images for use cases like product placement and assembly line checks. Explore Demo
Add Dense Captions to Images Generate human-readable captions for all important objects detected in your image. Explore Demo
Remove Backgrounds from Images Easily remove the background and preserve foreground elements. Explore Demo
Add Captions to Images Generate a human-readable sentence that describes the content of an image. Explore Demo
Detect Common Objects in Images Detect and extract bounding boxes for recognizable objects and living beings. Explore Demo
Extract Text from Images Use OCR to extract printed and handwritten text from images, PDFs, and TIFF files. Explore Demo
Extract Common Tags from Images Extract tags based on recognizable objects, scenery, and actions. Explore Demo
Create Smart-Cropped Images Automatically crop images to emphasize the most important areas. Explore Demo
Detect Faces in an Image Detect the location of human faces and their attributes in images. Explore Demo
Count People in an Area Analyze video to count the number of people in a designated zone. Explore Demo
Detect When People Cross a Line Detect when a person crosses a line in the camera’s field of view. Explore Demo

 


 


Language Studio


 


Language Processing Scenarios


 













































































Scenario Description Link
Extract PII Identify sensitive personally identifiable information (PII) in text. Explore Demo
Extract Key Phrases Quickly identify the main points from unstructured text. Explore Demo
Find Linked Entities Disambiguate the identity of entities found in text by linking to a knowledge base. Explore Demo
Extract Named Entities Identify and categorize entities in text using Named Entity Recognition (NER). Explore Demo
Extract Health Information Extract and label medical information from unstructured texts. Explore Demo
Analyze Sentiment and Opinions Provide sentiment labels and confidence scores at the sentence and document level. Explore Demo
Detect Language Determine the language used in the input document and return a confidence score. Explore Demo
Custom Text Classification Create custom text classification projects with labeled data and trained models. Explore Demo
Answer Questions Extract answers to questions from passages of text provided. Explore Demo
Conversational Language Understanding Projects Build projects with labeled data and trained models for understanding conversational language. Explore Demo
Orchestration Projects Build and manage projects that integrate multiple language services. Explore Demo
Summarize Information Produce summaries for conversations or documents using summarization APIs. Explore Demo


Document Translation

Batch translate documents into one or more languages either from local storage or Azure Blob Storage


 


 


Document Intelligence


 


Document Analysis Scenarios


 



























Scenario Description Link
Read Extract printed and handwritten texts along with barcodes and formulas from documents. Explore Demo
Layout Extract tables, checkboxes, and text from forms and images. Explore Demo
General Documents Extract key-value pairs and structure from any form or document. Explore Demo


 

Prebuilt Models Scenarios



 

































































Scenario


Description



Invoices


Extract invoice details including customer and vendor details, totals, and line items.



Receipts


Extract transaction details from receipts including date, merchant information, and totals.



Identity Documents


Extract details from passports and ID cards.



US Health Insurance Cards


Extract details from US health insurance cards.



US personal tax


Classify then extract information from documents containing any number of W2s, 1040s, 1098s and 1099s.



US mortgage


Extract information from a variety of mortgage



US pay stubs


Extract employee information, payment information including earnings, deductions, net pay and more.



US bank statements


Extract bank statements



US checks


Extract amount, date, pay to order MICR numbers, name and address of the player, and more.



Marriage Certificates


Extract details from marriage certificates.



Credit Cards


Extract details from credit cards including card number and cardholder name.



Contracts


Extract title and signatory parties’ information from contracts.



Business Cards


Extract contact details from business cards.


 


Gen-AI Safety Solutions


 



Safeguard your image content




 
























Scenario


Description


Link



Moderate image content


This is a tool for evaluating different content moderation scenarios. It takes into account various factors such as the type of content, the platform’s policies, and the potential impact on users. Run moderation tests on sample content. Use Configure filters to rerun and further fine tune the test results. Add specific terms to the block list that you want detect and act on.




Moderate Multimodal content


Run moderation test on image and text combined contents. Assess the test results with detected severities.










Safeguard your Text




 


































Scenario


Description


Link



Moderate text content


Run moderation tests on text contents. Assess the test results with detected severities.








Groundedness Detection


Groundedness detection detects ungroundedness generated by the large language models (LLMs).




Protected Material Detection


Detect and protect third-party text material in LLM modules.




Prompt Shield


Prompt shields provides a unified API that addresses the following types of attacks: Jailbreak attacks and Indirect attacks.






Real-time Safety Measures




 



















Scenario


Description


Link



Monitor Online Activity


This will display your API usage, moderation results, and their distributions per category. You can customize the severity threshold for each category to view the updated results and deploy the new threshold to your end. Additionally, you can edit the blocklist on this page to respond to any incidences.