There is LOW competition in public sector tenders. It's going to change

We all complain about HIGH COMPETITION in public sector tenders. 

Numbers tell a different story. The competition is in fact very LOW.

But things are changing - very fast.

Currently, 41% of tenders receive only 1 proposal. 

69% of tenders receive 3 proposals or less.

18% of tenders are canceled. Out of them, 6-13% due to insufficient bids, the rest for other or unspecified reasons (based on Hermix data for selected analyzed markets).

Between 2011 and 2021, the share of contracts receiving only one bid rose from 23.5% to 41.8%, and the average number of bidders fell from 5.7 to 3.2 (according to the European Court of Auditors). The median EU country has single bids for 28% of tenders (EC Scoreboard 2024).

The reality is that the public sector space is NOT crowded.

There are very few or no proposals submitted. There are a LOT of tenders canceled. A lot of these cancelation are caused by insufficient bids.

But things are going to change. Fast.

The new generation of AI tools allow companies to write and submit better, cheaper proposals.

Soon, companies will submit several times more tenders. Each tender will have hundreds or thousands or pages. Very well written, solid, verified, checked. Companies are usually not forbidden from using AI. And even if they would be forbidden, there are ways to go around such interdictions. 

Yes, they still need great technical proposal writers, architects, experts, bid managers. Without a great team, AI is not able to write a winning proposal. But with AI, humans can work so much faster, and better.

At the same time, most evaluators are not allowed to use AI, or have strong limits on AI use. Some are internal interdictions, some are public - e.g. for Eureka (Eurostars, Innowwide), or for Horizon grants (the research programme of the European Union).

This creates a huge disadvantage for public authorities and funding bodies. They use 30-year old tools and processes, to evaluate 100 times more, better proposals.

Authorities and funding bodies will need to adapt fast.

The evaluation criteria and tools will change sharply. 

We live incredible, exciting times.

We have no idea what the world will look like, even in 1 year from now.

Bibliography, further reading

AI has a place in research, but not in evaluation of Horizon Europe proposals, Commission says. (2024, March 21). Science|Business. https://sciencebusiness.net/news/ai/eu-publishes-guidelines-use-generative-ai-research

Eureka Network. (n.d.). Eurostars remote expert guidelines. Retrieved November 1, 2026, from https://www.eurekanetwork.org/wp-content/uploads/2026/01/remote-expert-guidelines.pdf

Eureka Network. (2024). Eureka Innowwide call 3: Evaluation guidelines (Version 1.0). https://eurekanetwork.org/innowwide-evaluation-guidelines

European Commission. (n.d.-a). Access to public procurement. Single Market and Competitiveness Scoreboard. Retrieved November 1, 2026, from https://single-market-scoreboard.ec.europa.eu/index.php/node/264_et

European Commission. (n.d.-b). Access to public procurement (Belgium). Single Market and Competitiveness Scoreboard. Retrieved November 1, 2026, from https://single-market-scoreboard.ec.europa.eu/node/373_hr

European Commission. (2023, December 4). European Court of Auditors (ECA) publishes special report on public procurement in the EU. Public Buyers Community. https://public-buyers-community.ec.europa.eu/news/european-court-auditors-eca-publishes-special-report-public-procurement-eu

European Commission. (2025). Standard briefing slides for experts: Horizon Europe (Version 13.0) [Presentation slides]. https://www.leuphana.de/fileadmin/user_upload/forschung/Files/transparenz/standard-briefing-slides-for-experts_he_en.pdf

European Court of Auditors. (2023). Public procurement in the EU: Less competition for contracts awarded for works, goods and services in the 10 years up to 2021 (Special Report No. 28/2023). https://www.eca.europa.eu/ECAHTMLNews/NEWS-SR-2023-28/en/body.html

European Court of Auditors. (2024, March 11). Special report 28/2023: Public procurement in the EU [Presentation slides]. Chaire EPPP, Sorbonne. https://www.chaire-eppp.org/wp-content/uploads/2024/03/Sorbonne_11.3.24.pdf

European Ombudsman. (2026). How the European Innovation Council and SMEs Executive Agency (EISMEA) and the European Commission ensure that the use of artificial intelligence (AI) by external experts is compatible with their tasks as evaluators of applications for Horizon funds [Case]. https://www.ombudsman.europa.eu/hu/case/en/70849

Luxinnovation. (2026, May 21). Key changes for Horizon Europe funding mechanisms. https://luxinnovation.lu/news/key-changes-for-horizon-europe-funding-mechanisms

Sáenz Pérez-Aradros, M. (2026, May 19). Simplification in Horizon Europe [Presentation slides]. European Commission, DG RTD, Common Implementation Centre. https://www.gov.si/assets/ministrstva/MVZI/Znanost/Obzorje-Evropa/Dogodki/2026-Dogodek-Obzorje-Evropa/Simplification-in-Horizon-Europe.pdf    

Scientific theories are True if they are Useful

Meta-modelism.

Science doesn't actually operate within a "true or false" paradigm. Rather, it works with scientific theories, which are models: simplifications of reality that help us understand one aspect of it at a time. Because reality is far too complex, chaotic, and analog, it is impossible to work with it in its entirety. So these models we work with, being simplifications of reality, a kind of "slices," are obviously "true" up to a certain point, in the sense that they correspond to certain aspects, beyond which they become false, no longer corresponding to other aspects.

Of course, a scientific theory must first and foremost be valid and reliable, but this isn't measured by whether it is true or false. It's measured by whether it is coherent, internally consistent, and repeatable.

Scientific theories, these "models," must simultaneously meet two contradictory goals: they must be simple, and therefore easy to work with, and at the same time complex, so as to simulate reality as closely as possible.

All right, then how do you measure the value of a scientific theory? How "true" is a theory? Is Euclidean geometry truer than non-Euclidean geometry? Is the theory of relativity truer, or string theory? Is the Big Bang model truer? Is the creationist theory true, the one where Adam was made from clay, belly button and all?

It doesn't matter so much how True a theory is.

A theory has to be functional. The right question is how useful a particular theory is in a particular context.

Microsoft is not replacing 27% of staff with AI

No, Microsoft is not firing 60 000 people because of AI. The press release is corporate marketing, interpreted incorrectly.

In fact, Microsoft only said that 60,000 employees are "eligible" for the buyout program. This doesn't mean all will take it. The real number will likely be 3-4 times less. Maybe 10-20k. Out of a total of 295000 employees. So no, Microsoft doesn't fire 27% of their workforce, and doesn't replace them with AI.

And even if Microsoft would fire 10-20k of employees: this would be not news, and it would not mean they would necessarily decrease their total workforce. 

Because they do this every year, before publishing their financial results.

Most big corporations announce these "cost reductions to increase efficiency". Shareholders like this, because it means higher profits and dividends. 

The interesting mechanism is that they fire 5-10% of staff, and hire others. 

Microsoft has announced layoffs almost every year for the past 10 years, with peaks in 2023 and 2025 (up to 7%). They announced 11000 in 2023, and 15000 in 2025, but still ended pretty much the same.

Over 2016–2025, Microsoft announced roughly 40,000–45,000 layoffs in total. Still, their staff grew from 116000 to 225000.

Yes, they always say it's for increasing efficiency and lowering costs. Now: with AI.

These redundancies often overlap with the 10% efficiency rule of Jack Welsh (you know, from the great GE. It's a mandatory case study in all MBA programs across the world). He said that the bottom 10%, the low performers, should always be fired, every year. And replaced with new, better employees.

Of course there are also companies who really reduce their workforce. Some even die. 

But don't believe every corporate press release. First look at their stock prices. 

The fear of AI is growing stronger

The fear of technology and AI is stronger than ever.

Asimov was one of the first to describe humans' fear of robots and AI.

Great stories, starting with Caves of Steel, The naked sun, and all the robot stories.

Asimov explained in many novels how humans would grow afraid of machines. And how machines can actually collaborate with humans.

The current general fear of AI reminds me of the irrational fear of robots described by Asimov 80 years ago.

80 years ago. 

We learned nothing. 

People faced with robots still behave like monkeys afraid of fire.

And the fear becomes stronger, as AI becomes stronger. 

AI wasn't Magic 2 years ago, but now starts to be Magic. We understand less and less how it works. And the results are more and more amazing.

So, we have these medievalists, as Asimov named them. People who refuse to use AI, and refuse technology. "We should ban AI for children. Ban mobile phones in schools. Ban Internet". Well, why don't you also ban Wikipedia, and Google. And dictionaries. And paper. And pencils. In the good old days, people were using their brain, and pergament, and hieroglyphs. And they were happy. Yeah, right.

But all this is nonsense. Everybody uses AI. Your Google Search uses AI. Google Translate. Google Maps and Waze. The voice recognition on your phone. Your apps that recognize songs and plants. Your health apps that track your heart beats.

We live in the most fascinating time in history.

The engineering view: AI is amazing, but not human. It's a tool. It's not perfect. It works

The engineering view: AI is amazing, but not human. It's a tool. It's not perfect. It works.

This text by Andrew Ng is so good, that I'm copying it here. It summarizes so well what we engineers have been saying for so many years.

<< The loudest voices stoking fears about AI dangers have made tremendous headway in the past two weeks. AI technology has not taken some unexpected, dangerous turn, but the hype around it — propelled by what appears to be a well orchestrated PR campaign — has drummed up considerable fear. I worry that it represents a setback for our field.

I have written frequently that fears of AI are overhyped. AI’s capabilities can be uncannily human-like and unpredictable, and it’s rational to worry when people who are directly involved express concerns. But I see the problems as a sign of the engineering work that ahead, rather than insurmountable barriers or the sky falling. AI technology continues to advance — which is a good thing! — but technical advances, poorly understood by the public, give those who seek to generate hype repeated opportunities to do so.

First, I don’t see any step up in the risk of human extinction from AI compared to a few months ago. The theories about this remain the same fantastical, science fiction scenarios as a few months ago. The biggest change in AI risk is its cybersecurity capabilities — a topic which we should take seriously — but this, too, will not lead to the end of the world.

The most notable recent event leading to increased fear was when an OpenAI team deployed an agent swarm that hacked into Hugging Face. Much of the popular press contained significant hype. For example, some publications reported that a swarm of 1,200 agents carried out the attack. While this was technically accurate, as I write this, I have about 1,300 processes running on my laptop. Yes, the ability to get large swarms of agents to work in parallel on a task is a significant technical advance, And, in computing, many processes run at the same time. So this shouldn’t be seen as some magical capability.

Additionally, OpenAI’s buggy sandboxing and monitoring processes were key to enabling this incident. Fixing these bugs and putting in place improved monitoring would be appropriate fixes, not pausing AI. There are many well known ways to attack software systems. The main advantage of AI agents is that they are relentless. They will tirelessly try many tactics — and have the patience to chain vulnerabilities together — that previously would have taken an infeasible amount of human effort. But in the long term, I believe the advantage will lie with defenders (because they have more information with which to identify bugs, which they can fix), but the cyber-threat landscape has changed significantly. There are still bottlenecks to identifying and exploiting a vulnerability. AI agents still have to try a lot of things to see what works, and taking these actions takes time and might be detected by defenders. This is why, even though it is now easy to obtain versions of leading open weight models that have had their guardrails removed or weakened, so they will not refuse to try to execute cyber attacks, the world has not ended.

I am also concerned about the anthropomorphization of AI in a lot of reporting, where LLMs and agents are unnecessarily treated as if they were people. If I wield a hammer, miss a nail, and accidentally dent the wall, it’s not the fault of the hammer. The problem lies in how I used the hammer. Similarly, if I prompt an agent and it hacks into someone else’s system, the responsibility lies with me, not the agent.

Of course, we want to build systems that are as safe and predictable as possible. (For example, an unsafe hammer would be one whose head randomly flies off under normal use.) Today’s agentic systems are not predictable, but I see no reason why, by applying sound engineering practices, we won’t be able to make them extremely safe to use. One new element in the forecasts of AI-enabled doom is AI companies disclaiming responsibility for their own products. “I didn’t do it; my out-of-control agent did!” There’s a balance to be struck between the responsibility of the tool maker and the tool user, but when something goes wrong, let’s hold the people building and/or using the hammer responsible, rather than the hammer. (By the way, if you’re worried about AI bioweapon risk, David Bellamy has a great post on why this, too, is overhyped. Briefly, the bottleneck in building a bioweapon is not intelligence, but lab work and manufacturing.)

Pausing AI progress will create much more harm than benefit. First, our adversaries will certainly not slow down. Second, engineering requires discovering problems empirically so we can fix them. If we pause AI by a decade, we will also delay finding and implementing safety engineering fixes by about the same duration.

Of course, the incentive to stoke fears — for regulatory capture, to garner attention, or to make one’s technology seem more powerful — remains the same as before. Disclaiming responsibility is a new one. Taking a hard technical look at the actual risks however, I see little factual basis for the degree of fear that’s been stoked up. We still have hard research and engineering work ahead to improve AI safety, but the beneficial applications continue to vastly outweigh the risks, and we should keep building. >>

Original text:

https://www.deeplearning.ai/the-batch/issue-371

Human-generated slop has a longer tradition than AI slop

Human-generated slop has a much longer tradition than AI slop.

We've generated huge amounts of bullshit, thousands of years. We published useless research papers, plagiarized essays, recycled books and PhD thesis, unreadable marketing PR, junk movies and music, unintelligible abstract or naive art.

And it's ok. This is how society functions. We are random, we copy, change, interpret, redesign, learn, and fail, continuously.

You have no right to be scandalized by AI slop.

We've lived with slop for ages.

Be amazed by AI achievements, and accept the AI slop. 

Just like we accept human-generated slop, and use and praise our achievements.

People start to hate AI

People start to hate AI.

It's a very interesting phenomenon. Stronger than anything we witnessed before.

This negative emotion is tangible - in social media, in communities, in legislation.

4 years ago, AI was a curiosity.

1 year ago, it was still interesting.

Now it is ubiquitous - so people started hating it, en masse.

The reason is the SPEED of the change.

AI is a storm. It doesn't need 100 years, or 10 years - like fire, automobiles, electricity, or email. It changes everything, now.

So, we are all scared.

And our reaction becomes irrational.

It's more irrational than people burning books, when we were afraid of new ideas.

More irrational than Samurais, Zulus, and Amerindians charging with arrows and spears, against rifles, railways, and machine guns.

Of course, people post-rationalize this irrational hate. It's a typical process: First we feel emotions, than we post-rationalize. And the post-rationalizations are usually exaggerated; extreme. AI is the Devil - or sometimes a God.

"Datacenters are killing the planet / consume all our electricity / water" (no, they are not. The energy consumption is quite small, compared to... everything else).

"AI is taking over our jobs" (well, yes, but it is also creating even more jobs).

"AI will set us free, and nobody will need to work again" (of course not) (corollary: let's tax AI / robots. Bill Gates keeps saying this, since 2017...)

"AI is stupid, and cannot replace humans / artists / lawyers etc" (this contradicts the previous argument, but well).

"AI cannot invent anything original" (well, maybe - but neither can you).

"AI will take over the world and kill us all" (no, it is not).

"AI doesn't know what it's doing, doesn't have awareness and intentionality" (again: this contradicts the previous argument, but well - who cares about logical contradictions in an irrational post-rationalization).

The key reason for this hate: the same as ever: fear of the new. Fear of technology. Fear of change.

Geniuses like Asimov, or Frank Herbert, forecasted this. They looked at history and arrived at the same conclusion: people hate change.

Asimov wrote about the C/Fe (carbon/iron) culture: impact of robots, rejection and hate of robots, good robots vs. bad robots, adaptation to C/Fe, changes in industry, economy, legislation; moral and ethical impact. 

So here we are. We will set barriers, impose legislation, block datacenters, and we'll set robots on fire.

And in the end, we will create the best brave new world.

Because AI is here to stay, and will change this world.