AskNews is positioned as a news layer and news API for AI, designed to enable AI systems to obtain trustworthy, structured news context. Its core selling point is the unified integration of licensed news sources and open web sources, and it provides more information enrichment and extraction than competitors (officially stated as 400% more), while real-time news access response times are under 100 milliseconds.
In terms of use cases, AskNews is aimed at AI systems and research scenarios that need news intelligence. The official description states that it is used by the UK Cabinet Office, frontier AI labs, and organizations in insurance, intelligence, private equity, and hedge funds, among others, to turn news into a basis for decision-making.
Main capabilities include: real-time search and large-scale retrieval; agentic deep research for complex questions and strategic analysis (DeepNews); real-time monitoring of topics, companies, and events (Alerts); historical news queries spanning years and decades; a knowledge graph showing connections among people, places, and events; and MCP/CLI access for agentic workflows. The official emphasis is that every data point comes with the context the model needs, rather than raw articles.
In terms of how to use it, AskNews provides Python and TypeScript SDKs as well as full documentation. The official command-line example given is: asknews research "Analyze how TSMC capacity limits are shaping OpenAI and Anthropic". The getting-started steps are three: choose a path (API, DeepNews, Alerts, or Forecasting), run a real query, and scale the model or upgrade the plan as traffic grows.
Regarding data sources, the official description says that paywalled news, the open web, X, Wikipedia, Google, Reddit, and podcasts can be accessed through a unified API without stitching together multiple integrations; DeepNews can conduct self-learning iterative research on sources such as paywalled news, open web content, X, Wikipedia, and others.
On pricing, the official listing includes several plans: a free start with 25 credits, 150,000 Lite tokens, and 150,000 Standard tokens (a card must be added, with no initial top-up required); a low-priced individual API plan (500 credits per month, 2M bonus tokens, 100 bonus credits, credit top-up rate as low as 20%); a plan for teams and advanced users (20,000 credits per month, 10M bonus tokens, 3,333 bonus credits, top-up rate 40% lower, doubled rate limits, exclusive tools X/Charts/Websearch); a professional API capacity plan (110,000 credits, 45M bonus tokens, 10,000 bonus credits, top-up rate 60% lower, exclusive tool Reddit); and an enterprise edition (commercial license, custom usage, support and deployment terms, as low as $0.0015/credit, rate limits up to 5000 RPS, enterprise support and SLA). Specific price figures are not fully provided.
Regarding limitations, the materials do not clearly state unsupported features or usage restrictions, and only mention that the free plan requires adding a card and that some advanced tools are exclusive to specific plans.
