CoGrader is an innovative AI-powered essay grading solution designed for educators that seamlessly integrates with platforms like Google Classroom and Canvas. With this tool, you can utilize your custom rubrics or create new ones through AI assistance, allowing you to deliver detailed and prompt feedback to your students at a speed that is 80% quicker, freeing up valuable time for you to spend with your children. This means you can focus more on what truly matters while ensuring your students receive the support they need to succeed.
Learn more
For developers, ARGOS Identity offers two APIs.
ARGOS ID Check (KYC/eKYC): launch the hosted Liveform from one URL in a browser or an iOS, Android or React Native WebView, or build your own capture UI on the Verify engine APIs. Your backend uses the REST API to fetch submissions, review pending cases, delete or partially delete data for GDPR and CCPA requests and download AML reports, and receives webhooks for created, submit, approved, rejected, updated, delete and AML result events. Each check reads the ID with OCR plus MRZ and PDF417 barcode checks, tests ID and selfie liveness, matches the face to the ID photo and screens the user against sanctions and PEP lists, with ongoing monitoring afterwards. Data is processed and stored in the region chosen for each project (US or South Korea), sensitive fields are encrypted with AES-256 at rest, and payload encryption is optional. A migration API imports existing verification records.
ARGOS Omni (Review Ops Agent): authenticate with an x-api-key header, create a profile under a workflow, upload files, text or JSON as items (PDF, DOCX, XLSX, PPTX, images, CSV, HTML, EML and more) and start an asynchronous analysis. The result follows the output schema you defined, with a verdict of verified, rejected or pending_review and the reasoning and source step for each extracted value, and every completed analysis can be downloaded as a PDF report.
Learn more
Lenz
Lenz is a robust API designed for audit-level fact-checking of AI-generated content, aimed at teams that require verification of factual assertions prior to presentation to clients or customers. This tool analyzes various forms of text, such as drafts, memos, or model answers, identifying factual claims that can be substantiated by credible public sources and providing a detailed outcome for each assertion. Its operational framework is built upon four key API functions: extracting claims that can be verified, performing a rapid multi-model assessment of these claims in bulk, elevating ambiguous statements to a comprehensive verification process, and facilitating follow-up inquiries that are based on the verification findings. The complete verification process is structured into five distinct stages: defining the issue, conducting research, engaging in debate, undergoing a panel review, and reaching a conclusion. Lenz meticulously searches through public information, evaluating evidence based on factors like authority, relevance, and timeliness, while also employing models from various vendors to present arguments for both sides of a claim. Ultimately, it delivers a verdict, a confidence score, references to cited sources, and a trace of the reasoning that allows users to conduct their own audits, ensuring transparency and reliability. With Lenz, teams can enhance their decision-making processes by ensuring all claims are thoroughly vetted before they reach their audience.
Learn more
ClaimBuster
ClaimBuster serves as the overarching platform for all fact-checking initiatives within the IDIR Lab. Originally launched to develop an AI system capable of identifying claims that warrant verification, it has progressively advanced towards achieving the ideal of fully automated fact-checking. While journalists primarily utilize ClaimBuster, it is designed for anyone motivated to confront misinformation. Accessing our models is straightforward through our API, which only requires the registration of a free API key. The success of ClaimBuster relies heavily on the contributions from individuals who label data. We encourage you to create an account and participate in labeling to enhance our model's effectiveness. Additionally, we've made our machine learning model training code available as open-source, inviting skilled AI engineers to contribute their expertise. Our claim-detection model even retweets posts it identifies as potentially needing fact-checking, expanding its utility in real-time. By participating, you can play an essential role in the ongoing battle against misinformation.
Learn more
