It has been about a week since our trip to Perugia for the International Journalism Festival 2026. As builders of a tool for journalists, we spent a lot of time at the sessions that gave us a glimpse of what tools practitioners actually recommend. We found many of them very interesting, while some are quite niche and would help only in specific use cases. We decided to share the ones that struck us as most broadly useful, and then tell you why we think epistimi fills a gap that none of them address.

The non‑Google tools below were demo'd by Craig Silverman at his session. Do visit indicator.media for more tools and his free weekly OSINT briefing.

1. Google Pinpoint

Pinpoint has been a powerhouse tool for journalists for some time. It was doing things that LLMs can do now, such as searching across large document sets, surfacing named entities, and finding connections, before LLMs were reliable technology. If you regularly deal with batches of documents and are not already using it, it is worth a look. It is free for verified journalists and researchers.

2. Google NotebookLM

NotebookLM has found quick popularity with anyone who deals with many documents or takes notes constantly. You upload files in different formats and extract information through a chat interface. Think of it as a very powerful Ctrl+F that also explains what it finds and always shows you the source.

Recent updates added web research, support for Sheets and Word documents, Drive file URLs, and even photos of handwritten notes. Genuinely useful for working through large sets of government documents or policy papers.

A word of caution: be careful about uploading sensitive or confidential documents to any AI‑powered tool. Google always emphasizes privacy and security, but the general advice is to use these with care. Once files are in the LLM dimension, it is hard to say exactly what happens to them.

3. SynthID

By this point you may think I am doing a Google advertising, but the tools they presented were the most powerful ones at the conference. SynthID, developed by Google DeepMind, embeds invisible watermarks in AI‑generated images at the point of creation. The watermark survives compression, resizing, and most common edits, making it detectable even after screenshots or re‑uploads. The limitation is that it currently only applies to images generated by Google's own tools, but the hope is this becomes an industry standard.

4. ImageWhisperer

Continuing with AI image detection: ImageWhisperer helps journalists figure out if an image is AI‑generated. It runs multiple checks and gives you clues and technical signals in a plain‑language verdict, but it is not a foolproof tool. Most journalists will recognise many of the individual signals, but having a structured report ensures you do not miss anything.

Important caveat from the tool itself: an authentic image does not equal true content. A real photograph can still show a staged scene. Always verify claims independently.

5. Wayback Machine Extension

The official Internet Archive browser extension is a commonly used tools in journalism. At its simplest, it saves any page you are viewing to the archive with one click. More usefully, it automatically detects 404 errors and offers you the last archived version of that page, which is essential when sources delete statements or official documents get quietly removed.

It also shows how many times a page has been archived and when it was last saved, lets you browse a full calendar of historical snapshots, and generates site maps and word clouds. For example, journalists can track changes to government websites and understand what has been the intent behind them. In essence, it is an indispensable tool to travel back in the URL time.

6. Ubikron

Ubikron is a browser‑based investigation workbench. The core idea is that it captures screenshots as you browse and organises them into projects, so your research stays structured rather than scattered across dozens of tabs and an unsorted folder of screen grabs.

Beyond screenshots, it extracts entities from pages such as emails, names, phone numbers, aliases, domains etc. It captures infinite‑scroll feeds, saves pages as PDF or MHTML for evidentiary purposes, and lets you add notes and annotations linked directly to clippings. A private server option is available for investigators who need full control over where their data is stored.

7. Yutori Scouts

Yutori Scouts deploys AI agents that monitor the web in the background for whatever you tell them to watch. You describe what you want tracked, whether a company, a political figure or a developing story, and the agents run continuously, notifying you when something relevant surfaces.

It can monitor specific URLs or the broader web, on a one‑off or recurring schedule, and delivers results by email digest or webhook. The system uses multiple sub‑agents in parallel: one for social media, one for academic sources, one for news, and so on. For journalists working on a beat this replaces the tedious work of manually checking sources every day.

8. OSINT Navigator by Indicator

OSINT Navigator is a project by Craig Silverman and journalist‑technologist Tom Vaillant. There are now nearly 7,500 digital journalism and research tools in existence, and finding the right one for a specific task is itself a challenge. Navigator solves that.

You type a natural‑language query, such as "How do I find the owner of a website?" or "How do I track a crypto transaction?", and it returns a ranked list of relevant tools from a curated database of nine independent OSINT toolkits. A notable feature for investigators working on sensitive stories: you can run Navigator with a local language model that processes entirely in your browser, with no API calls and no data leaving your device.

What We Did Not See at IJF26

After attending these sessions, one gap stood out. The tools covered AI, social media investigation, website analysis, and visual verification. Discovery was barely mentioned, and with search increasingly choked with AI‑generated noise, finding primary sources is harder than ever.

That's the problem epistimi is built to solve. We're aggregating official public records such as city council agendas, council minutes, municipal records, planning decisions, and other documents into one searchable place, built for local and investigative journalism. These documents are technically public, but in practice they are buried across hundreds of disconnected local authority websites, presented in inconsistent formats and, in some cases, not digitised at all. epistimi gives you access to official government records that no one else has aggregated.

If your work touches local government, planning, public spending, or municipal policy, epistimi.app is worth a look.