End to End Encryption, Nvidia, Layoffs & More: Tech Highlights March 2026
The Privacy Rollback & Tech Turbulence of Early 2026
Meta is pulling encryption off Instagram, the AI layoff wave is here in force, and Nvidia is worth more than the GDP of most countries. The last two weeks have been a lot. Let's break it all down.
How many of you actually live inside Instagram DMs? Sending voice notes at 2 AM, sharing memes, keeping a whole friendship alive through the little paper plane icon? Because if that's you, I have some news that directly affects how private those conversations actually are - or rather, how they're about to stop being private at all.
Beyond the encryption drama, the past fortnight has been genuinely eventful in the tech world. We had a massive AI-driven layoff wave, a chip company crossing market cap milestones no company has ever touched before, Google turning Chrome into something closer to a personal AI operator, and WhatsApp quietly opening its doors to people without accounts. This is your full breakdown.
Instagram Is Killing End-to-End Encryption - And It's More Complicated Than It Looks
This is the one that matters most, so let's give it the space it deserves. Meta has confirmed that end-to-end encrypted messaging on Instagram will no longer be supported after May 8, 2026. The feature, which ensured that only the sender and recipient could read a conversation - not even Meta - is being pulled. Users with affected chats are being prompted to download their data before the deadline.
What End-to-End Encryption Actually Does
Without getting too technical: end-to-end encryption (E2EE) assigns unique cryptographic keys to each device in a conversation. When you send a message, it gets locked using the recipient's key and can only be unlocked by them. No server in the middle - including Meta's - can read the contents. With E2EE gone from Instagram, your DMs will be readable by Meta's infrastructure. That means they can be scanned for content moderation, potentially used to inform ad targeting, and handed over to law enforcement under valid legal requests.
Meta's Explanation - And Why It Doesn't Fully Hold
Meta's official line is simple: "Very few people were opting in to end-to-end encrypted messaging in DMs, so we're removing this option." Low adoption, end of story. But that explanation collapses pretty quickly when you look at the history. Instagram only began testing E2EE in 2021 - five years after WhatsApp already had it as a default - and never finished rolling it out to the full user base. The feature was buried behind multiple taps and never advertised within the app. Meta essentially sabotaged the uptake, then cited low uptake as the reason to kill it. Several independent analyses have called this out directly as circular reasoning.
"This is the first time a major platform has ever rolled back encryption protections - and it's a worrisome sign for the future of private communications."
- PlatformerThe Real Reason: Regulatory Pressure and Timing
The timing makes the motivation clearer than any press release. The US Take It Down Act - which requires platforms to remove certain harmful content within 48 hours - takes effect on May 19, 2026. Instagram's encryption deadline is May 8: eleven days earlier. With E2EE active, Meta cannot proactively scan messages for violating content, because by design no one can read those messages. By removing it just before the law kicks in, Meta gets ahead of the compliance problem. Reuters also reported that Meta had internal warnings as far back as 2019 that encryption would "hinder the detection of illegal activities including child sexual abuse material" - so this tension was known for years.
TikTok made a similar announcement around the same time, publicly stating it has no plans to implement E2EE for direct messages. The company cited the same safety-vs-privacy tradeoff. When two of the world's largest messaging platforms move in the same direction in the same month, it is not a coincidence. It is a signal about where platform regulation is heading.
What This Means For Indian Users Specifically
India is one of Instagram's largest markets. With E2EE removed, message contents become accessible to Meta and obtainable by Indian authorities through legal requests - powers that India's Digital Personal Data Protection Rules 2025 (which came into effect in November 2025) explicitly grant the central government over data fiduciaries. Legal experts have flagged that Meta may need to update its consent framework for Indian users before May 8 to remain compliant.
Go to Settings → Your Activity → Download Your Information on Instagram and select your encrypted chats. If you want E2EE going forward, WhatsApp (still Meta-owned, but still encrypted by default) or Signal are your main alternatives. Meta's own guidance: "Anyone who wants end-to-end encrypted messaging can do that on WhatsApp." Make of that what you will.
India Consumes Data at a Historic Rate - And the Government May Tax It
India is not just a large internet market - it is the most data-hungry nation on the planet, and by a significant margin. According to TRAI data, the average Indian mobile user consumed 27.5 GB per month in 2025, nearly double the global average of roughly 15 GB. For context, monthly data usage in India was around 14-15 GB in 2022 - meaning it has almost doubled in three years.
What Is Driving This?
Several factors converge here. Reliance Jio's disruptive entry in 2016 permanently reset data prices in India, making it one of the cheapest per-GB markets globally. The 5G rollout - with Jio and Airtel leading aggressive network expansion - has enabled higher consumption in cities, while affordable 4G plans have extended access across rural areas. Short-video platforms are a significant driver: research suggests that around 40% of youth mobile data in India goes toward Instagram Reels and similar formats alone. Remote work, online education, and mobile gaming (BGMI, Free Fire) have added further layers of demand.
The ₹1 Per GB Tax Proposal
Against this backdrop, there are reports that the government is considering a ₹1 per GB levy on mobile data usage. This is still at a proposal stage and has not been formally tabled, but the implications are serious enough to think through. The average Indian user consuming 27.5 GB per month would face an additional charge of roughly ₹27.5 per month under such a scheme - which sounds small, but becomes meaningful when you consider India's massive base of 1.12 billion cellular connections. Critics argue such a tax would disproportionately impact lower-income users who have only recently gained digital access. Proponents suggest it could fund rural connectivity infrastructure. Nothing is confirmed - but it is worth watching.
The AI Layoff Wave Is Here: 45,000 Jobs Gone in Q1 2026
If you're in tech - or trying to get into tech - this one deserves your full attention. The first quarter of 2026 has seen a concentrated wave of workforce reductions unlike anything since the post-pandemic correction of 2023. Except this time, the nature of the cuts is structurally different. Companies are not just trimming overhiring; they are actively restructuring around AI.
Atlassian: 1,600 Roles, 10% of Workforce, All Framed as AI
On March 11, 2026, Atlassian - the Australian software giant behind Jira and Confluence - announced it was eliminating approximately 1,600 positions, around 10% of its global workforce. CEO Mike Cannon-Brookes framed it as a "self-funded investment in AI and enterprise sales." The majority of impacted employees were in North America (40%), followed by Australia (30%) and India (16%). The restructuring is expected to cost between $225 million and $236 million in severance and related charges, to be recorded by end of June 2026.
What makes this newsworthy beyond the numbers is the credibility gap. Just five months earlier, in October 2025, Cannon-Brookes publicly stated on a podcast that Atlassian would employ more engineers in five years, not fewer, and pledged to hire more fresh graduates in 2025 and 2026 than in previous years. The March 2026 cuts eliminated over 900 R&D roles - the exact departments he was promising to grow. Atlassian's stock rose 2% after the announcement was made, because Wall Street rewards headcount cuts regardless of the stated rationale.
"What we're seeing isn't just a correction - it's a restructuring. Companies are using AI as both a tool and an excuse to fundamentally reshape their workforces."
- Lacey Kaelani, CEO, MetaintroThe Broader Pattern: AI-Washing Layoffs
Atlassian is not alone. Block (the payments company led by Jack Dorsey) cut over 4,000 employees - nearly half its workforce - in February, citing AI automation. Amazon, Microsoft, and Meta have all made significant cuts in Q1. TechCrunch noted that enterprise-focused investors have begun predicting 2026 as the year AI starts taking a "meaningful toll on labour." The median re-employment time for laid-off tech workers has risen from 3.2 months in 2024 to 4.7 months in early 2026, reflecting the skills mismatch between eliminated and available roles.
The roles being cut are primarily in legacy support, middle management, and repetitive engineering work that AI can now handle or assist with at scale. Roles that are growing: AI product management, ML infrastructure, and engineers who can work alongside agentic AI systems. If you are currently job hunting in tech, skewing your portfolio toward AI-adjacent skills is not optional anymore - it is the filter most companies are applying at the hiring stage.
Your Next Phone Will Cost More - Here Is the RAM Price Problem
If you were waiting for mid-range flagships to get cheaper in 2026, that wait is getting longer. Brands like vivo and iQOO have flagged significant price increases in upcoming devices, and the culprit is not supply chain disruption or currency fluctuation in the traditional sense - it is the cost of RAM, which has spiked as demand for "AI phones" reshapes what counts as minimum viable memory.
Why RAM Prices Are Rising
The smartphone industry is in the middle of an AI integration race. Every major brand is shipping devices with on-device AI models for photography, translation, and assistant features. Those models require memory - and a lot of it. Whereas 8 GB of RAM was considered standard two years ago, 12 GB is now the new floor for devices targeting the AI phone segment, and premium flagships are pushing toward 16-24 GB. This surge in demand for high-bandwidth memory has driven up component costs, and manufacturers are passing those costs directly to consumers rather than absorbing them.
The broader semiconductor market has also tightened as Nvidia's data centre dominance and the broader AI infrastructure build-out compete for the same fabrication capacity as consumer memory chips. Analysts predict the high-RAM trend will continue through 2027 as more on-device AI workloads shift from cloud to local processing.
Gemini Is Now Living Inside Chrome and Maps - And It Is More Useful Than Expected
Google has spent the last six months watching AI-first browsers - from OpenAI's Atlas to Perplexity Comet to The Browser Company - try to eat into Chrome's market share by bundling AI assistants into the browsing experience. In late January 2026, Google responded with what might be its most significant Chrome update in years.
The New Chrome Sidebar
Gemini, previously floating in a separate window, now lives in a persistent side panel in Chrome - available on Windows, macOS, and Chromebook Plus. Built on Gemini 3, the update allows users to ask questions about any open page, compare multiple tabs simultaneously (useful for product research or reading across sources), and interface with a connected set of Google apps including Gmail, Calendar, YouTube, Maps, Google Shopping, and Google Flights. One example Google demoed: finding an old conference invitation in Gmail, checking Google Flights for the route, and drafting an email to colleagues - all without leaving the browser. The "auto browse" feature - where Gemini actually navigates websites and completes multi-step tasks on your behalf - is currently limited to US subscribers on the AI Pro and Ultra tiers, but the rollout has already begun expanding to India, Canada, and New Zealand as of March 2026, with auto browse coming later.
Gemini in Google Maps
Separately, Google expanded Gemini into Maps navigation for walking and cycling. While navigating on foot, you can now ask things like "What neighbourhood am I in?" or "Are there any cafes with a bathroom along my route?" without breaking stride or leaving the navigation screen. The update builds on the earlier Gemini-powered driving navigation feature that rolled out a few months prior. This is a notable shift in what Maps is being positioned as - not just a routing engine, but a conversational local guide.
WhatsApp Guest Chats: No Account, No Problem
WhatsApp has spent years operating behind a walled garden - to use it, you needed a verified phone number and an installed app. That is starting to change. The platform is testing Guest Chats, a feature that lets people join conversations via a web link without needing a WhatsApp account at all.
The immediate use case is obvious: businesses, event organisers, and group coordinators who deal with people who are not on WhatsApp. Instead of directing someone to download the app, you send them a link. They join, they participate, they leave. The implications for WhatsApp's reach are significant - especially in markets where the app has near-total penetration and non-users represent a shrinking but stubborn friction point. It also positions WhatsApp more directly against tools like Slack and Google Meet for lightweight professional coordination. No confirmed global rollout date yet, but testing is underway.
Nvidia Hits ~$4.4 Trillion. It Is the Most Valuable Company on Earth.
In November 2022, when OpenAI launched ChatGPT publicly, Nvidia was worth approximately $345 billion. As of March 2026, its market capitalisation sits at roughly $4.4 trillion - making it the most valuable publicly traded company in the world, ahead of Apple (~$3.77T), Alphabet (~$3.65T), and Microsoft (~$3T). That is a more than tenfold increase in value in under three and a half years.
Why Nvidia Specifically?
Nvidia does not make the AI models. It makes the GPUs - the specialised chips that train those models and run inference at scale. Every major AI lab (OpenAI, Anthropic, Google DeepMind, Meta AI, xAI) and every hyperscaler (Microsoft, Amazon, Google, Oracle) needs Nvidia's H100 and Blackwell chips to build and maintain their systems. The company is estimated to capture upward of 60% of big tech's AI infrastructure spending. Wall Street projects Nvidia could generate between $320 billion and $330 billion in data centre revenue in 2026 alone. That is more than the entire annual GDP of countries like Portugal or New Zealand.
"Nvidia was the first company to ever cross a $4 trillion market capitalisation. It did so faster than any company in history."
- Visual Capitalist, Feb 2026TSMC - the Taiwanese company that actually fabricates Nvidia's chips - now sits at a $2 trillion market cap, placing it sixth globally. Three of the top twenty most valuable companies in the world are AI semiconductor businesses. The chip industry, which was invisible to mainstream conversation five years ago, is now one of the most closely watched sectors on the planet.
The tech landscape has shifted considerably in two weeks, and not always in directions that favour regular users. Instagram is walking back a privacy feature while blaming users for not using it enough. Layoffs are accelerating, framed as AI investment but timed to market expectations. And the single most valuable company on earth now makes chips that power the AI systems enabling all of it.
A few questions worth sitting with: If platforms can quietly bury privacy features and then cite low adoption as a reason to remove them - what does meaningful user consent even look like? And as AI-justified layoffs become a normalized part of the corporate playbook, what does the re-skilling conversation actually need to look like for people entering the workforce right now?
What do you think - is the ₹1/GB proposal a reasonable way to fund digital infrastructure, or does it set a dangerous precedent for taxing internet access? And are you moving your conversations off Instagram after May 8? Let me know in the comments.
Comments
Post a Comment