Welcome to today’s briefing.

In this edition, we talk about an important topic. AI promised a four-day workweek, but instead what we’re seeing is a replacement by "vibe coding" sessions that run until forever.

In Albania, there is a legal battle brewing after the government used actress Anila Bisha’s likeness without her consent to create the virtual Diella, the AI Minister of the country.

And further down the content we talk about Replit’s massive 9 billion dollar valuation, and the release of the competitor of OpenClaw by Perplexity, a Mac Mini on the cloud.

Gartner predicts a major course correction is coming. Last weeks Signal newsletter, was focused around Anthropic’s prediction of which jobs AI will take first, which caused shockwaves accross readers. Now I’m turning into something else on this edition.

While the initial rush to integrate Generative AI led many companies to slash their customer service departments in search of efficiency, the data suggests a looming reality check. According to a new report by Gartner, half of the organizations that reduced staff in favor of AI-driven automation will be forced to rehire those human roles by 2027.

The issue isn't that AI is failing; it’s that it is hitting a complexity ceiling. While chatbots excel at handling routine queries, they struggle with high-stakes, emotionally charged, or nuanced customer issues. When these automated systems fail, the resulting customer effort spikes, leading to churn and brand damage.

Businesses are discovering that AI is most effective as a co-pilot rather than a replacement. The companies winning in the long term are those using AI to handle the volume, while empowering a leaner, highly-skilled human workforce to manage the critical touchpoints that drive loyalty.

Key Takeaway for Leaders: If you are cutting staff to hit short-term quarterly goals, you may be over-leveraging technology at the expense of your customer experience. The future isn't AI instead of humans, it’s AI augmenting humans to handle a more complex service landscape.

We were promised a four-day workweek, instead, we got vibe coding until 2 a.m. New research and evidence from the front lines of the AI transition suggest that rather than freeing up our time, generative AI is creating a compulsion loop that intensifies labor under the guise of creativity.

According to researchers at UC Berkeley, AI isn't just changing how we work—it’s expanding the definition of our jobs until the boundaries vanish entirely.

The Triple Threat of AI Work Intensification

AI makes specialized skills (coding, data analysis, graphic design) feel accessible. The result? Professionals are absorbing entire departments' worth of "ops" work into their daily routines, effectively doing two jobs for the price of one.

The "Slot Machine" Effect - Drawing on B.F. Skinner’s behavioral science, researchers have identified a variable-ratio reinforcement in prompting. Because the "perfect" AI response is unpredictable but feels attainable, users enter "epistemic rabbit holes," chasing one more prompt like a gambler at a slot machine.

FOBO (Fear of Becoming Obsolete) - A staggering 85% of employees are learning AI on their own time, driven by the anxiety that a weekend without "tinkering" puts them a week behind the rest of the world.

The Strategy for Leaders - For the C-suite, this looks like a productivity win. For the workforce, it’s a burnout time bomb. To maintain a sustainable edge, businesses must move from "accidental" AI use to intentional AI practices, protected focus windows, analog pauses, and a cultural rejection of the "1 a.m. breakthrough."

This guide from Deel breaks down how to build one global hiring system. You’ll learn about assessment frameworks that scale, how to do headcount planning across regions, and even intake processes that work everywhere. As HR pros know, hiring in one country is hard enough. So let this free global hiring guide give you the tools you need to avoid global hiring headaches.

Download the free guide today

Albania recently set a global precedent, but perhaps not the one it intended. What was meant to be a showcase of digital modernization has turned into a high-stakes legal battle over the right to own one’s own face and voice in the age of generative AI.

Last September, Prime Minister Edi Rama’s government unveiled "Diella" Albania's first AI-powered virtual minister tasked with overseeing public procurement. Framed as a tool to fight corruption and streamline EU integration, Diella was presented as a bold step forward. There was just one problem: Diella’s face and voice belonged to a real person, Albanian actress Anila Bisha.

Bisha is now suing the government for €1 million ($1.16M), alleging that her identity was repurposed without her consent.

The Scope of Consent: Bisha claims she agreed to lend her likeness to a digital assistant for the e-Albania services platform a routine chatbot.

The "Ministerial" Pivot: She argues she never authorized the state to turn her into a political figure or a virtual "minister" featured in official government communications.

Biometric Theft: The lawsuit asserts that her facial patterns and vocal data were used beyond the "specified, explicit, and legitimate purposes" required under GDPR-aligned data protection laws.

The Sovereignty of Likeness: This case exposes a critical blind spot in AI regulation. While we focus on bias and safety, we are neglecting synthetic identity appropriation. If a state can "borrow" a citizen's face for political messaging, the line between innovation and theft disappears.

The Permanent Avatar: Unlike a static photo, an AI avatar is code. It can generate unlimited speech and performances indefinitely. For Bisha, this isn't just a missed paycheck; it's a loss of autonomy over her own existence in the digital realm.

A Warning for Governments: As countries rush to adopt AI to "modernize" bureaucracy, the rush to be first often ignores basic human rights. Albania’s attempt to embody reform through AI has instead highlighted a lack of transparency.

The Bottom Line: AI may be efficient, but its power is borrowed from humans. If we cannot control our own faces and voices, we risk a future where our identities are scaled and redeployed by the state at the click of a button.

If you needed a signal that the AI coding gold rush has entered its hyper-growth phase, here it is. Replit has raised $400 million at a $9 billion valuation, tripling its price tag in just six months. The most mind blowing part isn't the valuation, it's the efficiency. Replit grew from $10M in revenue in 2024 to $240M in 2025. This isn't just a SaaS trend; it’s a fundamental shift in how value is created.

The term Vibe Coding, building software by simply describing the vibe or logic to an AI agent, is no longer a niche meme. It is now a multi-billion dollar sector.

Replit’s CEO, Amjad Masad is pivoting away from professional coders. His thesis? Software isn't technical work anymore; it's creative work. If you can describe it, you can build it.

The fresh capital will fund Agent 4, a tool designed to vibe code a startup from scratch, reportedly 10x faster than previous versions with a focus on real-time collaboration.

Replit, Cursor, and Lovable aren't just tools; they are the infrastructure for a new class of creator-entrepreneurs. The "gold rush" is real, and the traditional developer bottleneck is breaking.

The Vibe Economy:

Perplexity just threw down the gauntlet at its inaugural "Ask 2026" developer conference. The headline? They aren't just a search engine anymore, they want to be your entire operating system.

The company unveiled "Personal Computer" a software platform that turns a user-supplied Mac mini into a 24/7 autonomous AI agent. It’s a move that signals a massive shift in the industry: AI is moving from a "chatbot in a tab" to an "agent in your hardware".

Unlike standard AI tools that wait for your prompt, Personal Computer is designed to run continuously. By hosting the software on a Mac mini (which Perplexity calls the ideal "always-on host"), the AI gains persistent access to your local files, apps, and workflows, including: Slack, GitHub, Salesforce, and Notion, merging them with Perplexity’s cloud-based orchestration.

The Key Specs:

This isn't just a random choice. Following the viral success of the open-source "OpenClaw" project, the Mac mini has become the cult-favorite "brain" for AI enthusiasts. Perplexity is productizing this trend, offering a secure, sandboxed environment with a kill switch and full audit trails to soothe privacy concerns.

Perplexity isn't just targeting power users. They’ve launched Computer for Enterprise, claiming that internal tests saved $1.6 million in labor costs by completing three years worth of work in just four weeks. It connects directly to Snowflake and HubSpot, allowing teams to build financial models and dashboards via Slack without needing a data scientist.

Perplexity is betting $200 a month that you don't care which model you're using, as long as the "harness" is smart enough to use them all. By moving into local hardware, they are creating a moat that a simple browser extension can't touch.

The 2025 "Venture in Eastern Europe" report by How to Web has landed, and it paints a picture of a region in a high-stakes "recalibration" phase. While the global venture market has been on a rollercoaster, Eastern Europe has shown a unique kind of grit, even if the raw numbers suggest a flight to quality that makes life harder for average performers. Here are the critical takeaways:

In 2025, Eastern European startups raised a total of €3.6 billion. While the total volume remained stable compared to 2024, the deal count fell significantly. We are seeing a barbell effect:

The Giants: Large pools of capital are concentrating into a handful of "mega-rounds" (the top 10 countries accounted for nearly 90% of all capital).

The Graduates: Average round sizes for those who did get funded rose sharply, with pre-seed rounds up 45% and seed rounds up 43%.

If you aren't building with AI, you're fighting for scraps. AI dominated 43.2% of total funding in the first half of the year.

Enterprise Software led the charge with €720M in total investment, followed by Fintech (€522M).

New Frontiers: Investors are finally writing big checks for capital-intensive Deep Tech, with SpaceTech (€350M) and Defense/Security emerging as legitimate regional strengths.

Poland remains the undisputed heavyweight, attracting 20% of all VC capital in the region.

Ukraine showed incredible resilience, raising €163M despite the ongoing war, leading the "high-growth" group alongside Bulgaria and Slovakia.

Romania saw a 20% decline in volume (€103M), signaling a selective market where only high-conviction deals like Druid AI (€24.7M) and Digitail (€20M) are crossing the finish line.

The report highlights a "Diaspora Advantage." Founders with international experience, the ones who moved to London or the US and came back (or stayed connected), are the ones driving the biggest rounds. This ties directly back to our recent discussion on AI usage in Albania: exposure to global standards is the greatest predictor of local success.

Eastern Europe is no longer an emerging market experiment, it’s a global talent factory. However, the graduation gap is real. While we are great at starting companies, we still lack the massive growth-stage infrastructure to keep them here.

Thanks for making it to the end! I put my hard work and dedication into every email I send, I hope you are enjoying it.

Let me know your thoughts so I can make the next one even better!

Btw if you want to get your brand in front of a fast-growing audience of 500+ subscribers that are founders, investors, innovators, and tech professionals from South-East Europe all the way to Europe and the US, Signal connects the dots between local and global opportunities, and your message can be part of the story. Send an email at [email protected].

See you on the next edition,
Çelik