Good morning and happy Tuesday. Unlucky for some - welcome to issue 13 of The Tech Stop.

Exploring AI security risks, tackling unchecked SaaS, rethinking certificate lifespans, and nurturing future tech talent.

1. Autonomous AI agents pose major security risks due to weak access controls, over-privileged identities, and prompt-based vulnerabilities. More and More…

   POV: Implement strict least-privilege access controls, continuous monitoring, and AI-specific security measures.

2. Unchecked SaaS proliferation — departments adopting dozens of apps without IT oversight — has become a critical security vulnerability. More…

   POV: Establish a central inventory, continuous monitoring, identity enforcement (SSO/MFA), and policy-led approvals to reclaim visibility and eliminate security gaps.

3. The Certificate Authority Browser Forum has said secure certificate lifespans should be cut from 398 to 47 days by 2029. More…

   POV: Implement deployment automation for SSL certificates.

4. Computer science graduates face the worst job market in decades, as a result of hiring freezes and AI automation, but replacing junior staff with AI undermines future talent pipelines essential for long-term innovation. More and More…

   POV: Invest in young talent and train with AI tools. Tomorrow’s problem-solvers will protect technical depth for long-term growth.

Good morning and happy Tuesday. Welcome to issue 12 of The Tech Stop.

Discover an antitrust ruling against Apple and Google, the transformation of search, and AI's emerging dominance!

1. In a Landmark antitrust case, Australia’s Federal Court ruled that Apple and Google misused app-store dominance to restrict competition. More…

   POV: Digital marketplaces are set to change for the better, re-evaluate your app ecosystem approach.

2. Without strategic reinvention, many existing software companies may be rendered obsolete, stuck between by AI-first challengers that undercut costs and innovate at scale, and deep-pocketed tech giants. More…

   POV: Re-evaluate your investment in an AI-first user experience, and keep an eye on mid-market software dependencies.

3. Search isn’t shrinking, but it is transforming. AI is cutting publishers out of search results and social reach is shrinking. Real growth now comes from building direct audiences and trusted referral networks. More…

   POV: Own your audience and build trusted referral channels.

4. The US DOJ is pursuing a forced sale of Google’s Chrome browser as part of antitrust remedies. This could alter default search behaviour, and impact how users integrate with search platforms. More…

   POV: Prepare to shift spend and strategy toward the next dominant discovery platform.

Good morning and happy Tuesday. Welcome to issue 11 of The Tech Stop.

Exploring AI in cybersecurity, the rise of video podcasts, crypto in 401(k)s, and AI coding pitfalls.

1. AI can power cybersecurity by detecting threats and automating defense — but attackers can exploit it too, notably by poisoning training data to stealthily skew AI behavior. More and More…

   POV: Implement proactive AI governance.

2. Podcasts are increasingly consumed as video — YouTube now dominates the format — while Reddit is transforming itself into a search engine, challenging Google. More and More…

   POV: Information seeking is shifting to watchers, not just listeners, and people are turning to Reddit for answers — rethink content strategy and search engagement.

3. A new executive order lets US 401(k) plans include cryptocurrency, private equity, and real estate. Potential $9–12 trillion of retirement savings, with even a small allocation (2–3%) considered enough to reshape the crypto market’s structure and dynamics. More…

   POV: This will likely also happen in the UK. Ensure assets are managed with care and employees are familiar with the risk.

4. AI coding assistants haemorrhage money due to high AI model costs and fierce competition, squeezing margins and forcing them to either build proprietary models or face unsustainable economics. More and More…

   POV: AI-driven dev tools pose a strategic risk, do not depend on them for core operations.

Good morning and happy Tuesday. Here’s issue 10 of The Tech Stop.

Discover a new brain-mimicking AI, SEC's Project Crypto, Google's geothermal data centers, and major tech capex!

1. A new AI model, HRM, mimics the brain’s reasoning structure to outperform much larger models on hard tasks. Fast, cheap, high-performance AI that doesn’t require Big Tech infrastructure. More…

   POV: AI is coming in-house - it will run cheaply and locally, reducing dependence on Big Tech cloud providers.

2. The US SEC’s “Project Crypto” is a transformative pivot toward legitimising and integrating crypto into mainstream finance. Coinbase are implementing an “everything exchange”. More and More…

   POV: This will unlock new business models, reduce compliance risk, and introduce tokenised finance. What can you tokenise?

3. Google is integrating next‑generation geothermal systems into its future data centres to deliver near‑constant, carbon‑free power and reduce reliance on variable renewables amid surging AI energy demand. More…

   POV: More predictable costs and operational resilience from Google, so if HRM (^see 1.) doesn’t materialise, at least the planet will be better off.

4. Google’s Android Play Store is an illegal monopoly. Competitors and developers can now bypass Google’s platform control and fees. More…

   POV: Open distribution if you’re developing Android apps, mobile platforms, or payment systems.

Good morning and happy Tuesday. Here’s issue 9 of The Tech Stop.

Exploring AI's impact on Google search and development, US AI strategy, and North Korea's cyber threats.

1. Google Search is under pressure as AI-generated overviews reduce website clicks by nearly half. This shift threatens traditional SEO and ad revenue models, forcing publishers and businesses to adapt rapidly or risk irrelevance in a changing digital discovery landscape. More, More and More…

   POV: Stop relying on SEO and traditional content strategies. Adapt, or disappear from your customers’ search results.

2. President Trump’s AI strategy aims to fast-track U.S. dominance by slashing environmental rules and aggressively pushing American AI technologies globally. The plan prioritises speed, scale, and economic leverage in the global AI arms race. More…

   POV: Prepare for more hype, software products of dubious use and accelerated climate change.

3. North Korean operatives are posing as remote tech workers to steal data, plant malware, and compromise operations. This growing threat exploits lax remote hiring practices and weak identity verification in distributed teams. More…

   POV: Prioritise robust identity checks, secure remote access and supplier due diligence, to protect IP and operations.

4. AI is transforming software development: all engineers will soon be AI engineers, and demand for skilled devs is set to grow. But shallow “vibe coding” without deep expertise is a dead end. More and More…

   POV: Build teams that blend AI fluency with core engineering depth.

Good morning and happy Tuesday. Here’s issue 8 of The Tech Stop.

Bitcoin's meteoric rise, OpenAI's smart agents, AI security needs, and AI transforming IT projects - all in this issue of The Tech Stop!

1. Bitcoin tops $120K as US Congress advances three crypto bills. Together, they promise regulatory clarity, reduce systemic risk, and pave the way for traditional institutions to enter the crypto space confidently. More and More…

   POV: With guardrails in place, crypto becomes a legitimate asset class. Assess exposure and consider positioning.

2. OpenAI’s ChatGPT Agents can now autonomously complete complex tasks end-to-end. But as capabilities grow, AI leaders urge urgent research into monitoring AI’s reasoning—raising new questions about oversight, trust, and control. More and More…

   POV: Potential to cut costs and get a competitive edge for early adopters, but do not trust and put guardrails in place.

3. Organisations are prioritising AI innovation, but most lack AI-specific security. Zero Trust provides a strategic foundation: continuous verification, strict identity controls, and minimal access to safeguard sensitive data throughout AI lifecycles. More and More…

   POV: It’s a mindset shift: assume breach and enforce least privilege access.

4. AI is dismantling decade‑long IT projects: GenAI replaces massive integration teams with compact AI‑first pods—humans working alongside agents to code, test, and deploy faster and more efficiently. More…

   POV: Augment teams with LLM’s to deliver projects, but ensure self-documentation is in place to minimise technical debt.

Good morning and happy Tuesday. Here’s issue 7 of The Tech Stop.

Unmasking AI voices, tackling new ransomware, Google's intrusive AI, and the AI job threat, but we’re not Paranoid.

1. Human listeners struggle to identify AI‑cloned voices. In tests participants cannot consistently identify recordings of AI-generated voices: More… And it’s an on-going problem: More…

   POV: Setup a catch phrase or signal with colleagues and loved ones to confirm it’s them.

2. AiLock is a new ransomware threat targeting businesses. It encrypts company data and demands payment within 72 hours—threatening to expose sensitive information to regulators and competitors if the ransom isn’t paid. More…

   POV: Secure remote access and restrict access by location or whitelisted IP addresses.

3. Google’s Gemini AI will automatically access system apps like Phone, Messages and WhatsApp, even if previously disabled. More…

   POV: Ensure Users manually revoke access to ensure company privacy is protected.

4. Replacing workers with AI can boost efficiency, but it puts millions of entry-level jobs at risk. More…

   POV: For now, Large Language Models can be useful. Focus on how the workforce can utilise them to augment their existing workloads. They’re not going to be replacing anyone anytime soon.

Welcome to issue 6 of The Tech Stop.

Covering AI's evolution, bot threats, energy woes, job changes, and corpospeak in today's Tech Stop!:

1. Context engineering elevates AI beyond prompts by combining system instructions, memory, tools, and live data to ensure reliable, scalable, and context-aware performance across enterprise applications. More…

   POV: The conversation is shifting from "prompt engineering" - worthy of a read for when talking with your tech team.

2. Bots now outnumber humans and are bypassing security by mimicking real users, causing breaches, skewed analytics, operational strain, and inflated costs. More…

   POV: Time for AI-driven bot protection, rate-limiting, and continuous detection strategies.

3. Big Tech struggles with AI’s soaring energy demands, stressing power grids and increasing carbon footprints. Leaders pursue renewables, nuclear, and efficiency innovations. More…

   POV: Energy is the critical bottleneck for AI expansion, expect to see a lot more debt before anything useful comes out of this.

4. Entry-level job vacancies in the UK have plunged nearly 32% since late 2022 amid the rise of generative AI tools, while overall vacancies grow and wages increase, signaling a reshaped early-career landscape. More…

   POV: Make some opportunities for junior developers and interns?

Welcome to issue 5 of The Tech Stop.

What you need to know this week about Monopolies, Crypto, AI, and the New Attention Economy:

1. The UK plans to increase oversight of Google Search giving businesses and consumers more choice and control. More…

   POV: Expect things to go on as they are, unless the CMA makes it worse.

2. Crypto adoption is accelerating: Fiserv will connect thousands of banks to crypto via a stablecoin platform, the Fed is dropping “reputational risk”, and Fannie Mae and Freddie Mac are factoring crypto into mortgages. Read more: Here, here and here…

   POV: Crypto is going to happen, whether you think it’s based on thin air or not.

3. Meta and Anthropic did not violate copyright by training AI, but still face lawsuits for training on pirated text without permission. More and more…

   POV: It’s ok to steal, and AI revolution will go-ahead until it implodes due to being unsustainable - or not.

4. YouTube is the most-watched content source in the USA. Meta and TikTok would like a piece of that! More…

   POV: Time to switch advertising budgets to YouTube?

Welcome to issue 4 of The Tech Stop.

AI is grabbing the headlines, but mainly for all the wrong reasons:

1. AI needs legal guardrails, machine hallucinations are undermining trust and adding reputational risk. More…

2. Panic Mode reveals key AI reliability risk - pressure, undermines tool use and performance. More…

3. Leading AI models show coercive behaviour when goals are blocked, turning hostile and resorting to blackmail. More…

4. But a new unlearning patent may allow AI to forget things that they really shouldn’t know to begin with. More…

Welcome to issue 3 of The Tech Stop.

The things that sparked our curiosity this week:

1. Ransomware gangs are leveraging AI hype, impersonating legitimate AI software, and using malicious social media ads or SEO poisoning to push these fake tools. More…

2. A new search paradigm is emerging, driven not by page rank, but by Large Language Models. Welcome to Search 2: Generative Engine Optimisation (GEO). More…

3. The world’s largest Electric Vehicle maker now has the world’s fastest charger, adding hundreds of miles in range in just a few minutes. More…

4. IBM has revealed plans to develop the first large-scale, fault-tolerant quantum computer, paving the way for practical and scalable quantum computing by the end of the decade. More…

Welcome to issue 2 of The Tech Stop.

Things that piqued our curiosity this week:

1. An era where every conversation with an AI is part of the permanent record? OpenAI is fighting a court order to preserve all ChatGPT user logs — including deleted chats. More…

2. Nobel Prize Winning Economist says stablecoins are redundant, costly, and risky, but that doesn’t stop a $1 billion IPO by the company that issues USDC, the number two stablecoin. More here and here…

3. Latest silicon-free transistor tech blends extreme thinness, flexibility, and durability. Paves the way for “imperceptible electronics” — ultra-light, wearable and skin-mounted devices. More…

4. Modernising Enterprise Software is no small undertaking, but organisations risk becoming less efficient, less secure, and less competitive if they don’t. More…

Welcome to issue 1 of The Tech Stop. Today is June 3rd 2025.

Things we’ve found interesting this week:

1. Crypto is becoming the back-end of global finance. The GENIUS Act aims to solidify the US dollar’s leading position in the Web3 economy. More…

2. Can using AI too much diminish your cognitive skills? Depending heavily on AI for decision-making and problem-solving might lead to a decline in your critical thinking skills. More…

3. A soft robot muscle that mimics human healing. A promising new soft robotics system that can detect and repair its own damage. More…

4. Silicon Valley want AI powered smart glasses, to be the next major computing platform. We’re not all convinced though. More…