💰 Perplexity Makes $34.5B Bid for Chrome
AI search startup Perplexity has offered $34.5B to buy Google’s Chrome, amid an antitrust case that could force Google to divest the browser. Valued at $18B, Perplexity pledges to keep Google as default search, invest $3B in Chromium, and back Chrome for 100 months. Analysts call the move bold but unlikely to succeed.
🤝 Anthropic Offers Claude to U.S. Government for $1
Anthropic will give its Claude for Enterprise and Claude for Government tools to all three U.S. government branches for $1 per agency for a year, mirroring OpenAI’s recent offer. Partnering with the GSA, it will also provide technical support. The move reflects fierce competition among AI firms to win federal contracts and strengthen ties with policymakers.
🔥 Claude Now Processes Entire Software Projects in One Go
Anthropic’s Claude Sonnet 4 can now handle 1M tokens—enough for full software codebases or dozens of research papers—in a single request. Available via API and Amazon Bedrock, it delivers perfect recall in tests and new capabilities for complex workflows. The leap intensifies competition with OpenAI and Google while raising pricing and safety stakes.
🧮 Huawei Unveils Algorithm to Ease Chip Reliance
Huawei has introduced Unified Cache Manager, a software tool that speeds AI model inference by up to 90% and boosts throughput 22-fold by optimizing data use across HBM, DRAM, and SSD storage. Set to be open-sourced in September, it could help China reduce dependence on foreign HBM chips, a market dominated by SK Hynix, Samsung, and Micron.
Enjoying our newsletter? Forward it to a colleague—
it’s one of the best ways to support us.
☝️ POWERED BY KORBIT AI
Code Faster, Ship Smarter
Korbit’s AI powered code review platform helps engineering teams catch bugs, boost code quality, and keep projects moving — all with instant, context-aware feedback on every pull request.
Korbit delivers:
Automated PR summaries so reviewers spend time fixing, not guessing
High signal reviews tailored to your codebase and standards
Actionable team insights on review velocity, compliance, and quality trends
👾 FORWARD FUTURE ORIGINAL
The Future Is Hybrid: Mastering the Art of Human–AI Collaboration
The idea whose time has come is not the replacement of human intelligence by artificial intelligence, but their strategic integration. As AI capabilities and usage surge across industries, the fundamental question facing professionals, particularly non-technical professionals today isn't whether AI will transform their work—it's whether they'll be active participants in that transformation or casualties of their own psychological resistance to change. → Read the full article here.
🚀 STARTUPS
Will AI Power the First One-Person Unicorn?
Generative AI is enabling founders to build and operate companies almost entirely alone, fueling talk of billion-dollar “one-person” startups. Incubators like Audos provide capital and AI agents that handle product design, marketing, and back-office work, letting non-technical founders such as Solace’s Sarah Gwilliam scale without staff. Supporters hail this as a new wave of entrepreneurial “democratization.”
But challenges loom. AI tools still make costly mistakes, business ideas are easier to copy, and most infrastructure is controlled by tech giants who may also compete for the same markets. With hiring delays and lean teams becoming a badge of honor, the race to create the first solo-built unicorn is on—though it may be as much about navigating dependencies as seizing opportunity. → Read the full article here.
🤖 MODELS
GPT-5 Pushes AI Forward, But Keeps the Revolution Waiting
OpenAI has unveiled GPT-5, a “unified” model that blends the strengths of its previous AI systems into one, capable of selecting the best approach for any task. The upgrade delivers the strongest performance yet in coding, scientific reasoning, and knowledge work, edging out its predecessors but falling short of a transformative leap. Analysts see it as two months of progress in line with long-term trends rather than a sudden breakthrough.
Still, those trends point toward potentially world-changing AI within a few years. METR’s data shows capabilities doubling every ~200 days, putting full workday-level automation within reach by 2027—and possibly superintelligent systems soon after. That has safety advocates urging tighter regulation, warning that OpenAI’s rapid pace risks edging into self-improving AI. GPT-5 may be incremental, but its trajectory keeps the AGI debate very much alive. → Read the full article here.
⚛️ SCIENCE
Meta Spots Early Signs of Self-Improving AI, Tightens Grip on Most Powerful Models
Meta CEO Mark Zuckerberg says the company’s AI systems have begun improving themselves without human input—a development he calls the first step toward artificial superintelligence (ASI). In a July 30 policy paper, he argued that ASI could far surpass human cognition and accelerate discoveries “not imaginable today,” but warned that such technology must be released with caution. Future high-capability models, he said, will not be fully open-sourced.
The announcement follows academic work on self-modifying AI, including a “Gödel Agent” framework that rewrote its own code and proved each change beneficial, outperforming human-designed agents in coding, math, and reasoning. While Zuckerberg remains optimistic about AI as a tool for personal empowerment, he acknowledged that exponential self-improvement could reshape humanity’s trajectory—and that controlling access is now part of Meta’s strategy. → Read the full paper here.
🛰️ NEWS
What Else is Happening
⚠️ China Warns Against NVIDIA H20: Beijing is urging local tech giants to favor domestic GPUs over NVIDIA’s H20, risking a major hit to US export deal profits despite recent licensing breakthrough.
📋 AI Speeds Police Paperwork: Departments nationwide are testing Axon’s Draft One, cutting report writing time by up to 70%—but critics warn of bias, accuracy, and transparency risks.
🧐 Musk to Sue Apple Over ChatGPT Favoritism: Elon Musk accuses Apple of rigging App Store rankings to boost OpenAI, calling it an antitrust violation and vowing immediate legal action.
🚫 Reddit Blocks Internet Archive: Citing AI firms’ policy violations and privacy concerns, Reddit now limits Wayback Machine access, likely aiming to protect data and boost licensing revenue.
📱 Pixel 10 Set for AI-Powered Debut: Google’s Aug. 20 event will unveil Pixel 10 lineup with Gemini AI upgrades, G5 chips, major camera boosts, and new foldable, watch, and earbuds.
💼 JOB BOARD
Now Hiring: Handshake, Avahi, & More
📽️ VIDEO
The Industry Reacts to OpenAI's Open Source!
ICYMI: OpenAI’s new open-source GPT OSS models wow industry with strong performance, low training costs, and local-run capability, sparking strategic and safety debates.
That’s a Wrap!
📢 Want to advertise with Forward Future? Reach 550K+ AI enthusiasts, tech leaders, and decision-makers. Let’s talk—just reply to this email.
Thanks for reading today’s newsletter—see you next time!
Matthew Berman & The Forward Future Team
🧑🚀 🧑🚀 🧑🚀 🧑🚀
