AI Weekly: GPT-5 Arrives, a Fierce Price War, and India's AI Surge

Hi everyone,

This week was marked by a series of seismic shifts that are fundamentally reshaping the AI industry. The long-anticipated arrival of OpenAI's GPT-5 and Anthropic's Claude Opus 4.1 set a new benchmark for reasoning, while a new open-source model from China's Zhipu AI ignited a fierce price war, threatening to commoditize the core technology that powers the AI revolution.

Concurrently, a wave of scientific breakthroughs demonstrated AI's power to solve humanity's most complex challenges. These global trends found a unique echo in India, where a surge in funding, new product launches, and government initiatives underscored the nation's strategy to become a sovereign AI powerhouse.

Let's dive into the biggest stories.

The New Frontier: Foundational Models & Breakthroughs

The race for the most capable AI model heated up with major new releases, while a new front opened up in the battle for market share: price.

• OpenAI Launches GPT-5

  OpenAI officially released GPT-5 globally on August 7, replacing all previous versions. This new iteration shows a 40% improvement over GPT-4 on complex tasks and seamlessly handles text, images, and voice in a single platform. Planned "mini" and "nano" versions promise to bring this power to everyday devices.

  • Use Case: A fintech company integrates GPT-5 into its customer support chatbot. The bot can now analyze complex financial queries, process image scans of documents for verification, and provide secure, voice-based financial advice, cutting average resolution time by 25%.

  • Domain: Financial Technology (Fintech) / Customer Service/ Content/ Development

• Anthropic Rolls Out Claude Opus 4.1

  Anthropic released Claude Opus 4.1 on August 5, delivering a powerful upgrade focused on best-in-class coding, reasoning, and agentic task completion. It is quickly becoming a favorite for developers building reliable enterprise automation and analytics tools due to its precision in handling complex problems.

  • Use Case: An e-commerce company uses Claude 4.1 agents to autonomously handle inventory forecasting and customer returns. The AI analyzes sales data, predicts stock needs, and processes returns by interacting with logistics software.

  • Domain: E-commerce / Operations Management

• Tencent Debuts Open-Source Hunyuan AI Models

  Tencent unveiled a new family of open-source Hunyuan models designed for a wide range of environments, from edge devices like smartphones to enterprise servers. These compact models democratize access to sophisticated AI, allowing for more affordable and private on-device applications.

  • Use Case: A consumer electronics brand uses a Hunyuan model in its new smart speaker to power voice commands locally, ensuring user privacy and smooth operation even without an internet connection.

  • Domain: Consumer Electronics / IoT

AI in the Real World: Science, Medicine, and Your Daily Apps

AI's potential to solve complex challenges and integrate into our daily lives was on full display this week.

• AI-Designed CRISPR Enzyme Redefines Biotechnology

  In a landmark achievement, startup Profluent Bio created "OpenCRISPR-1," the first genome-editing enzyme designed entirely by generative AI. This breakthrough enables faster, safer, and more customized gene editing, opening a new frontier in therapeutic development by creating a novel and highly effective protein.

  • Use Case: Medical researchers use OpenCRISPR-1 to perform tailored gene edits for treating rare genetic disorders, enabling precision medicine at an unprecedented scale.

  • Domain: Biotechnology / Medical Research

• China Opens World’s First AI Hospital Simulation

  China launched a virtual AI hospital where digital doctors, powered by AI agents, can manage up to 3,000 simulated patients a day. The platform was developed to assist in diagnostics and medical training, providing a risk-free environment to address healthcare workforce shortages.

  • Use Case: A rural medical college uses the AI hospital platform for student training, allowing learners to diagnose and manage a wide variety of simulated cases and gain near-real clinical experience.

  • Domain: Healthcare / Medical Education

• Microsoft Edge Adds Copilot Mode

  Microsoft officially launched "Copilot Mode" in the Edge browser, an experimental feature that transforms the browser into an intelligent assistant with contextual awareness across all open tabs. With permission, Copilot can analyze and synthesize information from multiple web pages at once to answer complex questions.

  • Use Case: A home buyer with multiple property listings open in Edge can ask Copilot, "Based on these tabs, which property is in my budget, has three bedrooms, and is in the best school district?" Copilot parses the data from all open tabs to provide a single, synthesized answer.

  • Domain: Personal Productivity / Real Estate

• Google's NotebookLM Rolls Out Video Overviews

  Google's AI research assistant, NotebookLM, rolled out its "Video Overviews" feature on August 6. The tool automatically generates a narrated video presentation from a collection of source documents you upload, synthesizing key information into a slide deck with relevant visuals.

  • Use Case: A university student uploads all their lecture notes and research papers into NotebookLM to create a concise, narrated video summary for each major topic, making exam revision more engaging.

  • Domain: Education / Research

The Creative Revolution Continues

The toolset for artists, designers, and developers received a major upgrade with new AI-powered features focused on control and consistency.

• Runway & Luma Labs Pioneer "Modify with Instructions"

  The generative video space leaped forward as Runway (with Runway Aleph) and Luma Labs (with Dream Machine) both introduced in-context editing. These tools allow creators to modify existing video with simple text prompts like "change the character's jacket to a traditional Indian kurta," turning the AI into a collaborative post-production assistant.

  • Use Case: A marketing agency uses a single source video to create localized ads for different countries by issuing text commands to change backgrounds and products, avoiding expensive reshoots.

  • Domain: Marketing & Advertising / Video Production

• Ideogram & Midjourney Add Character and Narrative Control

  Ideogram and Midjourney released features to tackle the critical challenge of consistency. Ideogram's "Character" feature maintains a character's appearance across multiple images from a single reference photo, while Midjourney's start/end frame control gives creators power over an animation's flow.

  • Use Case: A graphic novelist uses Ideogram's Character feature to generate all their illustrations, ensuring their protagonist has the exact same appearance in every scene.

  • Domain: Illustration / Publishing / Animation

• Adobe Enhances Photoshop with New Firefly-Powered Features

  Adobe rolled out a significant Photoshop update on August 7, introducing "Harmonize" and "Generative Upscale". Harmonize automatically blends a new object into an image by matching the lighting and shadows, while Generative Upscale increases image resolution up to 8 megapixels without losing clarity.

  • Use Case: A marketing designer uses Harmonize to place a studio photo of an armchair into dozens of different lifestyle photos, with Photoshop's AI automatically adjusting the chair's lighting to match each new environment.

  • Domain: Graphic Design / E-commerce Marketing

Special Focus: India's AI Ecosystem Heats Up

It was a big week for AI in India, with a surge in funding, new "AI for Bharat" product launches, and proactive government initiatives underscoring the nation's focused strategy.

• Government of India Expands AIKosh Data Platform

  On August 8, the Indian government significantly expanded its AIKosh platform, which now provides startups and researchers with access to over 1,200 India-specific datasets and 217 AI models. This initiative aims to fuel indigenous AI innovation by providing the high-quality, vernacular data needed to build solutions tailored for the Indian population.

  • Use Case: A health-tech startup in India uses a large, anonymized dataset of medical images from AIKosh to train a diagnostic AI model specifically tuned to the genetic and environmental factors of the local population.

  • Domain: Health-Tech / Public Sector / AI Research

• Indian AI Startups See Major Funding Rebound

  The week of August 4-9 marked a major funding rebound for India's startups, raising $205.31 million (a 57% weekly increase), with AI-focused companies as a major driver. A report on August 10 also revealed that India-origin AI startups are securing significant early-stage funding from top-tier global investors.

  • Use Case: DPDZero, an AI-powered debt collection platform, was among the companies that secured funding this week to enhance its AI algorithms and expand its services to more financial institutions across India.

  • Domain: Fintech / Venture Capital

• New "AI for Bharat" Tools Launch

  On August 6, MakeMyTrip unveiled a multilingual GenAI trip planning assistant that allows users to plan and book complex travel itineraries through natural conversation in both Hindi and English. On August 7, Fractal announced its agentic AI platform, Cogentiq, is now available on the AWS Marketplace to help organizations build intelligent agents for customer engagement.

  • Use Case: A family more comfortable conversing in Hindi can now use the MakeMyTrip app to plan a multi-destination pilgrimage by stating their request via voice, with the AI agent building and booking the entire itinerary.

  • Domain: Travel & Tourism / E-commerce

• Karnataka Government to Assess AI's Impact on Workforce

  On August 5, the government of Karnataka announced a comprehensive survey to evaluate AI's impact on the state's workforce. The assessment will provide crucial data on how AI is reshaping employment, which jobs are at risk, and what new skills are in demand.

  • Use Case: Based on the survey's findings, a local university could redesign its computer science curriculum, reducing focus on manual tasks now automated by AI and adding new courses on AI ethics and agentic workflow design.

  • Domain: Public Policy / Education / Human Resources

The pace of change is accelerating, and the ability to direct these powerful new tools is becoming the most critical skill for any professional.

To help you master this new era, the 2025 edition of my book, "Prompt DOT AI: The Art of Writing Generative AI Prompts," is the completely updated playbook. This isn't just a list of prompts; it's a strategic guide to becoming an "AI Orchestrator," commanding a powerful suite of specialized AI agents to create anything you can imagine.

This edition is packed with new content for 2025:

• Master the AI Titans: In-depth guides for the top models: OpenAI's ChatGPT, Google's Veo 3, Kling, DeepSeek, and more.

• Direct AI Films: A complete guide to AI video creation, from storyboarding and consistent characters to advanced lip-sync and sound.

• NEW! Sculpt in 3D: A brand-new chapter on the revolutionary frontier of AI 3D model generation from text.

• NEW! JSON Prompting: Gain ultimate control over complex video generation with structured, blueprint-style prompts.

• Advanced Creative Workflows: Discover how to create consistent characters with LoRA models, compose music by uploading your own melodies to SUNO, and build complex projects with multi-tool pipelines.

• Actionable Formulas: Easy-to-use prompt recipes for social media, brand strategy, and more, plus a real-world music video case study.

Stop just using AI—start directing it. Grab your copy today and master the art of AI in 2025!

Stay creative,

Da Sachin Sharma