This collection holds 50 frontier signals from November 2026. Frontier means new (2025–2026), surprising and resting on a concrete fact with a primary source rather than a speaker’s opinion. We add no more than 50 signals a month: fewer, but ones that change the picture. December signals will be a separate collection.
The signals are arranged by 20 trends and the four Decoder worlds: CS, the autopilot world; CE, the rules world; DS, the autonomy world; DE, the community world.
Where the signals come from:
research and data: Stanford Digital Economy Lab, Art Basel and UBS, Gallup, Edelman, Reuters/Ipsos, Hugging Face, OpenAI and Upwork, WHO and others;
futures institutes’ reports: what CIFS, FTSG, IFTF, Dubai Future Foundation, RAND, McKinsey, Gartner, WEF, VML and the kleinkleinklein trend review added and dropped over the year, one signal per institute (material for the lecture “The Future Has Changed Hands”, 11 November 2026);
events and cases: news and decisions by companies, governments and communities, each linked to a primary source;
the III International Symposium “Creating the Future” (Moscow, 12–13 November 2026): only projects and studies that surprise in their own right. A separate note on the symposium will follow.
Assigning signals to worlds is our reading and can be challenged. Collection date: 20 November 2026.
The signals are grouped into 20 trends, each with its own card in Trends. The “Collection trend” link on a signal card leads there.
Each signal is a separate note. Below is the catalog: search and filter by world, life domain, theme and macrotrend.
According to Hugging Face, open-weight Chinese models took 41% of downloads versus 36.5% for American ones; about 80% of US AI startups use Chinese models somewhere.
ADP data through June 2026 show that employment of young workers in highly AI-exposed occupations is about 19% below where it would be had it kept pace with peers.
According to Revelio Labs, middle-management vacancies have fallen 40% since 2022; according to Gallup, a manager had 5 reports in 2013 and 12 in 2026.
The winning project of the film pitch at the “Creating the Future” symposium: people who lost their jobs to neural networks open a paid service of live human contact.
Vogue describes neurodiagnostics, neurofeedback and brain-optimisation programmes in private clinics; some week-long programmes cost up to 30,000 Swiss francs.
In Future 100 for 2026 (15,639 respondents, 16 markets) VML replaced escapism and “delulu” with “disoptimism”: acknowledging darkness without surrendering to it.
39% of Gen Z collectors show their collection in person only to family and close friends, almost twice as many as other generations; 42% share it online only in closed groups.
The Token Tax study of 10 models and 16 African languages found that inefficient tokenisation makes texts in low-resource languages cost more and come out worse.
MGX (Mubadala and G42) closed a fund of about $49 billion, invested in OpenAI, Anthropic and xAI, bought Aligned Data Centers with BlackRock and is building an AI campus of up to 3 GW near Paris.
In McKinsey’s Technology Trends Outlook Web3 is no longer on the open list, AI has split into infrastructure, agents and science, and energy, networks, chips and data centres are described as constraints on growth.
The Gulf and China buy and lease farmland abroad: the UAE holds about 960,000 hectares, and Africa has about 60% of the world’s uncultivated arable land.
China’s generative-AI safety standard requires a test bank of at least 2,000 questions; models are checked for alignment with “core socialist values” before release.
The Future Today Institute, now FTSG, replaced its annual Tech Trends Report of hundreds of trends with Convergence Outlook 2026: 10 nodes where technologies converge, including “Compute Shock”, “Emotional Outsourcing”, “The Corporate Panopticon”, “Agentic Economy” and “The New Labor Equation”.
The kleinkleinklein META 2026 review combined 70+ reports and 400+ trends: the words of the year are “anxiety” and “overload”, and 74% of strategists say there are too many empty trends.
In September 2025 the “Youths Against Corruption” server gathered more than 10,000 people to debate and vote for an interim prime minister; on 12 September Sushila Karki took office.
At the 2024–2025 protests in South Korea, concert lightsticks, fan chants and online mobilisation became a form of political expression: fans carried practices of collective voting into politics.
The Institute for the Future in Palo Alto has shifted the centre of gravity of its forecasts to trust and authenticity: “the loneliness business is booming”, “discernment becomes the most essential soft skill”, “the human verification crisis” and a rebuilt system for trusting science.
In Chicago parents are challenging the fact that AI Overviews cannot be switched off in Google’s school search; school council members discussed a three-year moratorium.
At the “Creating the Future” symposium, HSE researchers said they had analysed about 500 works of science fiction and identified three groups of competencies: clarity of thought, frames and responsibility, trust and humanity.
In Global 50 2026 the Dubai Future Foundation moved “Energy Boundaries” from megatrends into “assumptions” and replaced “Saving Ecosystems” with “Evolving Ecosystems”: from protecting nature to adapting to a nature already changed.
The Global Risks Report 2026 names geoeconomic confrontation as the most likely trigger of a crisis, with disinformation and polarisation at the top of short-term risks.