The Global Centre for Risk and Innovation (GCRI)
Global systems reports try to describe risks that no single discipline owns: climate, food, finance, health and technology acting on each other. Synthesis at that scale is where small weaknesses in the evidence base become large ones, because the report is read as a summary of everything known. This page sets out how global syntheses are built and reviewed, where their evidence is thin or skewed, and how their statements travel into decisions.
GCRI · Established · service level R0
A global risk assessment is a synthesis of evidence on risks that cross disciplines and borders, such as climate, food, finance, health and technology acting on each other. At that scale small weaknesses in the evidence base become large, because the report is read as a summary of everything known; uneven geographic and language coverage is a common one.
Where people and networks concentrate, the exposure through which shocks travel between systems, shown for the contiguous United States as an example geography for this subject.
The working screen behind this subject: its structure, terms and states. Names are placeholders and values illustrative.
Global systems series · Annual synthesis · Edition 5
| Annual synthesis | Global | Data check | 89 | W38 |
| Country profile 7 | Country | Peer review | 98 | W18 |
| Water outlook | Sectoral | Drafting | 47 | W16 |
| Methods note 4 | Methods | Released | 100 | done |
Interdependent global system risks. The dossier below sets out how the risk develops, what is measured, where the data come from, how it is modelled and who decides.
Published literature is concentrated in some regions, languages and disciplines. A synthesis that assesses only what is indexed in English inherits those gaps and can present an absence of evidence as an absence of risk.
Combining findings across fields with different methods and definitions can produce a statement that no single study supports. Calibrated language and traceable accounts keep the level of agreement visible instead of averaging it away.
A headline from a summary is quoted in briefings, media and other reports, often without its confidence qualifier or its scope. Each repetition moves the statement further from the evidence behind it.
A flawed or retracted source cited in a global synthesis can be carried into national plans and sector studies that cite the synthesis rather than the source, so a correction upstream may not reach users downstream.
Each term as this page uses it. Where a standard-setting body defines the term, the wording follows that body.
| Term | Definition |
|---|---|
| Confidence | In IPCC usage, a qualitative judgement of the validity of a finding based on evidence and agreement. |
| Likelihood | In IPCC usage, a probabilistic statement of an outcome, such as likely for a probability above two in three. |
| Literature cut-off | The date after which new studies are not assessed in a report. |
Each indicator carries its unit, its cadence, the kind of source behind it and the evidence label a value would carry. No values are shown: a programme supplies them under its own authority.
| Indicator | Unit | Cadence | Source type | Label |
|---|---|---|---|---|
| Assessed literature by region of author affiliation | percent of cited works by region | per report | Bibliographic metadata matched to institution registry | ESTIMATED |
| Assessed literature by language of publication | percent of cited works by language | per report | Reference list metadata | REPORTED |
| Key statements carrying calibrated confidence or likelihood | percent of summary statements | per report | Editorial audit of the summary | REPORTED |
| Review comments received and answered | count received; percent with recorded response | per review round | Review comment log | OFFICIAL |
| Literature cut-off to publication interval | months | per report | Report front matter | DECLARED |
| Cited works with a persistent identifier | percent of references | per version | Reference metadata | OBSERVED |
| Cited works later retracted or flagged | count per report | quarterly | Retraction records matched to references | OBSERVED |
The topic as a computational graph: named sources feed the methods, an evidence review gates the outputs, and the decisions stay with the authorities that hold them.
Fig. 02Nodes carry their type, two metric rows and a state; connectors show what feeds what; dashed frames name the owners; the timeline is the cadence. Design of the pipeline, not a record of a run.
Named public datasets and institutional records, each with its licence class. Sources by agreement are used only under their terms and never republished.
| Source | What it gives | Licence class |
|---|---|---|
| IPCC report archive and review records | Assessment reports, summaries for policymakers, expert and government review comments with author responses | Free to read, IPCC copyright, reuse terms per IPCC |
| IPBES assessment archive | Global and regional biodiversity assessments, summaries and review documentation | Free to read, reuse terms per IPBES |
| UNDRR Global Assessment Report archive | Global assessments of disaster and systemic risk | Free to read, United Nations terms |
| OpenAlex | Works, authors, institutions, languages and citation links across disciplines | CC0 |
| Crossref metadata | DOIs, references, funding and update notices for scholarly works | Open metadata, no reuse restrictions |
| Retraction Watch Database via Crossref | Retractions, expressions of concern and corrections with reasons | Openly available through Crossref, attribution requested |
Nexus Reports prepares evidence. The decisions below are taken by the authorities named, under their own rules, and never by the platform.
| Who decides | The question | Evidence this page serves |
|---|---|---|
| Ministries of foreign affairs, finance and environment | Which global risks warrant national attention, joint action or negotiation positions | Summary statements with calibrated confidence, the scope of each and its evidence gaps |
| Intergovernmental assessment bodies and their working groups | Where a later assessment should focus or where the literature is too thin to assess | Coverage profiles by region, language and discipline |
| Research funders | Which gaps in the evidence base justify new calls or data collection | Mapped gaps where confidence is low because evidence is limited rather than because it disagrees |
| Report editors and review editors | Whether every comment has been answered and every key statement can be traced | Comment logs, traceable accounts and reference integrity checks |
Global risks rarely respect the boundaries of disciplines or ministries. A heatwave strains power grids, cuts crop yields, raises mortality and moves commodity prices at once. Studying such risks one sector at a time misses the connections through which they amplify. The series exists to describe those connections with evidence and to say plainly where knowledge runs out.
Three series share the global view, each with a different time horizon.
Annual analysis of interdependent global system risks and the resilience of the systems that carry them.
Systems whose failure would have strategic consequences, such as critical infrastructure, digital networks and supply chains.
Long-horizon outlooks and horizon scans, labelled as foresight and never as prediction.
Research on systemic risk has shown that tightly coupled systems with little slack can turn a local failure into a widespread one, and that the same efficiency that lowers costs in normal times can remove the buffers needed in a crisis. The series looks for these patterns: shared dependencies, concentration, synchronized shocks and feedback loops. It also looks for the opposite, the redundancy, diversity and institutional capacity that let systems absorb shocks and recover.
Dependency maps and network analysis identify couplings. They depend on data that are often incomplete for private and informal flows.
Trend and correlation analysis, built on probability and statistics, describes what has happened. Past relationships may not hold under new conditions.
Coupled models and scenarios explore what could happen. Each carries stated assumptions and uncertainty assessed with uncertainty quantification and model validation.
Structured elicitation fills gaps where data are absent. It is labelled as expert judgement and its panel and process are described.
The current service level is R0: information and enquiries. An enquiry receives a reference and names the next step; applications, payments, compute allocations and registrations open only when their team, terms and review capacity exist.