Thomson Reuters Launches Its Own LLM
September 14, 2026
Thomson Reuters is the large news, content, and technology company. It uses the word “trust” on many of its outputs. They firm now has it own smart large language model called… wait for it … Thomson. “Thomson Reuters Leverages Its World-Class Data Assets To Launch Its Own Frontier Model” controls this model, and it may reduce the firm’s need to hire lawyers to do certain lawyerly things like read cases, mark them up, and generate an annotation. My hunch is that non-lawyers at Thomson will give Thomson a spin in order to create Thomson style outputs. Is this too many Thomsons? Of course not. Lord Thomson of Fleet is thrilled with the name “Thomson.”
Instead of spending billions of dollars and waiting years to build infrastructure, Thomson Reuters used an open source foundation and invested $40 million to train its LLM on the most important information-centric task Thomson employees undertake. Because Thomson Reuters used open source software, Thomson’s hands have Thomson by the scruff of the neck. The undertaking appears to have sidestepped the costs incurred by some other AI pioneers.
Here is what makes Thomson different from other frontier models:
“Thomson starts from a strong open-source foundation. What makes it different is what happens next: state-of-the-art mid-training and post-training techniques, drawing on decades of authoritative content from Westlaw, Practical Law, Checkpoint, and Reuters, with hundreds of subject matter experts integrated from the design of training objectives through to the final evaluations…The model has been trained on less than 10% of Thomson Reuters content so far, and what comes next is not simply feeding it more data. It is continued discovery of new kinds of specialization and understanding, made possible only by building on decades of proprietary content and editorial expertise.”
Is Thomson a leader or a follower in AI? What will be Reed Elsevier’s response? What will the lawyers now employed to do the old-fashioned mark up react to having Thomson a click away on their computer screen? Will Thomson output information that makes the occasional error (just like real lawyers)?
None of these issues are important in the grand scheme of AI deployment. What’s important is [a] cost cutting, [b] efficiency, [c] marketing, and [d] showing subscribers that Thomson Reuters is definitely moving down the information highway with AI as a copilot (yeah, lame pun, but I will blame my dinobaby boss).
Whitney Grace, September 14, 2026
Comments
One Response to “Thomson Reuters Launches Its Own LLM”
Got something to say?
-
- Subscribe to Beyond Search
Feature archive
News archive
-
Youg tee sex tubeDirty adult emailsGay bracesBoy and mom pornoWaterproof dildo harnessMelissa rycroft boobMilitary porn tubeAmateur facials uk torrent xnxx Large mature nipplesCasual sex encounters sucking dickAdult protective services omaha neFucking littly babiesJapan pussy fuckingBest of biazzar free pornLinda
blair sex videoWolff’s hardcore https://xnxxmega.com/ Amy reid
analSmall boob thaiAnime pussy cumAlbuquerque amateur photo xxxFree taboo kay parker porn galleriesTeens drinking driving
againstMiley gets nakedBeauty hairy very xnxx Xxx naked disney charactersBlow jobs are heart healthyOver her leg eye on spankSizzling
sex instruction videosFree costume porn videosSolo ebony with big
tits youpornWomen bodybuilders pussyTranssexual escort ivyMilf first lesbian fuck free videoMilf hunter + colleenJust pictures of huge black cocksCunnilingus top
powered by vbulletinAmater naked girlsNude singaporeansRodney’s north huntingdon pa gayClit hand job tubeBbw veronica vancouver washingtonCasey cooper flashes boobsCute sexy e-card animated