Skip to content

Topic

Enterprise AI Assistants and Search

Company-wide AI coworker and agentic search platforms grounded in internal organizational knowledge.

Current stories

buildOne report1 publisher

Workday lets TypeSafe's Jev choose which design-system components Sana puts on screen

Workday gave TypeSafe's Jev every atomic component in its design system and lets the model pick what Sana shows, said chief AI officer Joel Hellermark. TypeSafe claims 70 to 500 milliseconds per answer; whether Workday sees that depends on how many screens reduce to closed questions.

Publishers:thenewstack.io

Reality

Evidence48
Adoption35
Hype gap+25
Incentives65
Confidence55
buildConfirmed6 publishers

Claude Dashboards links every number to the SQL that produced it

Anthropic's Claude Dashboards, now in beta on paid plans, shows the SQL query behind any number a user clicks. The New Stack reported that Anthropic has not yet said whose credentials run each refresh, a detail data teams need before connecting a production warehouse.

Perspective Coverage

7 publishers
Builder
Builder 36%
Operator
Operator 40%
Investor
Investor 24%

Reality

Evidence70
Adoption
Not scored
Hype gap+10
Incentives60
Confidence68
investOne report1 publisher

Microsoft folds its consumer Copilot into one assistant built for corporate buyers

Microsoft is merging its consumer and workplace Copilot assistants into one product for corporate customers after a six-month rebuild, Bloomberg reported. It leaves personal chatbots to OpenAI, Google and Meta and ties Microsoft's AI bet to the corporate accounts it already serves.

Reality

Evidence40
Adoption
Not scored
Hype gap+15
Incentives
Not scored
Confidence45
buildOne report1 publisher

AWS wraps five quality layers around the LLM assistant used by over 4,000 of its leaders

AWS says NarrateAI, its data assistant for over 4,000 executive leaders, reaches about 99 percent numerical accuracy through five layered techniques. All five sit outside the model, so copying the design means extra AWS accounts and parallel evaluators on every paragraph.

Publishers:aws.amazon.com

Reality

Evidence35
Adoption30
Hype gap+20
Incentives75
Confidence40
productOne report1 publisher

IT buyers get a unified Microsoft Copilot app with usage billing still marked 'evolving'

Microsoft's unified Copilot app adds Code and Autopilot tabs for Microsoft 365 customers, with usage-based billing under a cost model still called 'evolving'. Firms deep in Microsoft 365 get AI that works inside their own tenant, on a bill they cannot yet forecast.

Publishers:zdnet.com

Reality

Evidence35
Adoption
Not scored
Hype gap+30
Incentives75
Confidence40
productOne report1 publisher

Amazon's cure for shadow AI ships on the desktop from outside its European sovereign cloud

AWS has made the Quick desktop assistant generally available on macOS and Windows and pitched it to IT as the sanctioned answer to shadow AI. It runs in Frankfurt, London and Ireland, and not in the Brandenburg sovereign cloud.

Publishers:thenextweb.com

Reality

Evidence55
Adoption34
Hype gap+28
Incentives74
Confidence55
buildOne report1 publisher

Amazon pitches the Quick desktop app to IT as containment for shadow AI

The argument AWS makes to IT for its newly general desktop assistant is that unapproved tools already move data out of controlled environments, and its answer is audit trails in CloudWatch and CloudTrail plus four named certifications.

Publishers:aws.amazon.com

Reality

Evidence26
Adoption24
Hype gap+46
Incentives93
Confidence62
leadershipOne report1 publisher

Manufacturers funded AI faster than they defined what it was supposed to change

The gap between Rockwell's 95% of manufacturers funding AI and McKinsey's 2% embedding it in plant operations gets easier to read once you notice the 60% who never set a target for what the deployment should achieve.

Publishers:forbes.com

Reality

Evidence40
Adoption32
Hype gap+18
Incentives76
Confidence52
buildOne report1 publisher

Stripe paid $7B for a router. Glean is selling the choice not to call a model at all

Glean says its customers mostly turn on automatic model selection to control spend, not to improve answers. The routing layer, not the model, is where enterprise AI budgets now get decided.

Publishers:latent.space

Reality

Evidence32
Adoption64
Hype gap+34
Incentives82
Confidence44