Gameplify aims to hone AI skills of BD professionals: Moshiur Rahman

DO: What is your main goal behind Gameplify's AI initiatives? What problems do you want to solve?
Moshiur: My goal is to help Bangladeshi professionals use AI in everyday work safely, not just experiment with chatbots. Gameplify provides our community foundation; Xyz Cloud Bangladesh Limited carries the training, AI and cloud offerings.

We are addressing three gaps: practical skills, control over sensitive documents and affordable access to tools and computing. The direction is connected: learn a workflow, practice in Xyz Cloud AI Lab, keep project materials in Xyz Drives, and eventually run AI-assisted work and heavier processing from that workspace.

DO: Which AI projects are you currently prioritizing, and what progress have you made so far?
Moshiur: We are prioritizing the Applied AI Accelerator for Professionals, Xyz Cloud AI Lab and Xyz Drives.

The Accelerator combines two days of hands-on training in Dhaka with 60 days of guided practice. Faculty include Dr. Md Saiful Alam Chowdhury (University of Dhaka), Maliha Tabassum (BUP) and Ariful Islam Arman (DhakaPost.com). The first open cohort has been rescheduled to December 2026.

The Lab provides structured exercises, workflow templates and metered AI access; Drives supplies the storage foundation. This year, we trained around 500 Dhaka Reporters Unity members across seven batches, 50-plus staff of a leading online news outlet, and 60 journalists in a two-day program.

DO: How can your initiatives help turn AI from discussion into practical applications in Bangladesh?
Moshiur: We should measure completed work, accuracy and time saved, not certificates alone. Our method is Brief → Context → Limitations → Draft → Verification → Refinement, taught through participants’ own work in a Bangla-English mix, priced in taka.

For the planned Drives�"AI integration, a journalist could select interview recordings and documents, ask questions across those sources or request a draft with source references, then save the reviewed result alongside the originals. A business could build a report from approved project files. The goal is less repeated uploading, traceable sources and human review�"not an AI system browsing every private folder.

DO: Are you working on AI training and skill-development programs? Who are you mainly targeting �" students, young professionals or entrepreneurs?
Moshiur: We focus on working professionals and entrepreneurs, final-year students and recent graduates, and institutions such as newsrooms, corporate teams and universities.

Professionals practice on real tasks; students build work samples, without a job guarantee. Institutions can develop shared workflows and rules for handling sensitive information.

We are also designing a 12-month Scholars Track with monthly challenges, quarterly assessments and merit-based educational grants toward accredited U.S. online master’s degrees. We remain independent of universities; admission decisions are theirs.

DO: How can AI help journalists in their daily work? What skills should journalists develop to use AI effectively?
Moshiur: AI can help transcribe interviews, summarize documents, translate between Bangla and English, analyze spreadsheets and search archives. Journalists still need to define the task, check quotations and numbers against originals, protect sources and follow newsroom disclosure policies.

The essential skills are clear briefing, verification, data literacy and editorial judgment. Confidential material belongs only in approved workflows with understood access, processing and retention terms. Encryption in storage does not automatically make external AI processing private.

DO: How can journalists and users identify AI-generated misinformation, deepfakes and fabricated content?
Moshiur: Start with the original source: who published it, when and in what context? Reverse-search images and video keyframes; corroborate through independent reporting or direct contact.

C2PA Content Credentials can help establish recorded origin and editing history, not whether a claim is true. Watermark checks such as SynthID have limited coverage. Missing credentials or a negative detector result prove neither authenticity nor fabrication.

Visual and audio anomalies are clues, not verdicts. Before sharing, pause and check Bangladeshi fact-checkers such as Rumor Scanner and Dismislab.

DO: Beyond AI, what other technologies is Gameplify working on? In which countries are you currently operating besides Bangladesh?
Moshiur: Gameplify runs real-time quizzes and trivia with automated scoring and live leaderboards in Bangladesh and the Seattle area.

Xyz Drives is our privacy-first storage platform, with user-controlled encrypted vaults, encrypted vault-to-vault transfer and regional storage choices. Its secure-vault technology is U.S. patent-pending.

We are preparing AI-enabled Drives and cloud-compute plans, coming soon for professionals, students, creators and teams. Plans would package storage, metered AI access and cloud computers for coding, analysis and media processing without buying powerful hardware. Customers would authorize selected files, with outputs saved back to Drives. Processing location and retention terms must be explicit; storage-region choice alone is not a guarantee about AI processing.

Plan limits, pricing and availability will be announced separately. Interested customers can contact Xyz Cloud Bangladesh for launch updates. We operate from Seattle and through Xyz Cloud Bangladesh Limited in Dhaka.

DO: You have worked in Silicon Valley and at major tech companies. What are the tech giants planning for the future of AI, and how should Bangladesh prepare to keep pace?
Moshiur: My Amazon and Oracle Cloud work was in Seattle. The direction I see is toward AI agents that complete multi-step tasks, backed by computing infrastructure and controls over data access.

Bangladesh needs practical skills, affordable compute, secure data handling, strong Bangla-language resources and funded implementation of national AI policy. Our role is to connect those needs at a usable scale: training, tools, storage and computing.

Bangladesh does not have to build every frontier model. It should build the capability to use AI responsibly and productively.