Whether you have questions, need a solution, or want to explore our services, TrueSolv is here to help.
Home>>News>>Dreamforce 2026 Announcements, Full Recap
Dreamforce 2026 Announcements, Full Recap
By: Anastasiia Rashkina
September 14, 2026
More than 50,000 people came to Moscone Center, and over 122,000 more watched online. 🎤 Now the same question lands on every leadership team: what from all of that actually changes our Salesforce this year? We sorted the biggest Dreamforce 2026 announcements by the one thing that matters for planning — when they reach your org.
The big picture in one paragraph
Salesforce built the whole event around becoming an Agentic Enterprise and drew a clear four-layer stack. Data 360 holds data, metadata and memory. Customer 360 carries application logic. Agentforce runs the agents. On top sits the brand new AIforce, an interface layer that brings Salesforce data and actions into the tools people already use, starting with Slack and Claude, while keeping every existing permission and business rule in place.
Four layers, one architecture, reaching into the tools you already use.
Available right now ✅
🔀
Multi Agent Orchestration is generally available, so several agents can pass work to each other instead of acting alone.
🤖
Most of the new job ready agents are generally available, including Piper for inbound leads, Casey and Fin for service and customer experience, Carter for commerce, Marshall for back office and supply chain, and Paige for employee HR and IT requests.
🧠
More model choice in Agentforce arrived with Amazon Bedrock models, plus Anthropic and NVIDIA models, available to Agentforce customers.
🗄
Data 360 zero copy now reaches more AWS sources, including Apache Iceberg tables, Amazon Aurora, Amazon RDS and SageMaker Lakehouse.
📊
Salesforce and Tableau inside Gemini Enterprise are live for Google Cloud customers.
📖
The Well Architected Framework now has five pillars with an agentic lens, and it is probably the most practical free read of the week for architects.
In beta or pilot 🧪
💬
Salesforce in Claude is in beta on all paid Claude plans.
🧬
Koa, the first Salesforce CRM reasoning model built together with NVIDIA, is piloting with select customers. General availability is expected in winter 2026 for US regions.
🔌
Headless 360 MCP Server is in open beta, while the Data 360 MCP Server is already generally available.
🆕
AIforce is officially announced, and Salesforce is still finalizing pricing and packaging.
Coming this fall 📅
⚙️
Agent Optimizer is planned for October.
📣
Campaign Agent is expected in Marketing Cloud Next Advanced by October 2026.
🎯
Hunter, the outbound pipeline agent, is planned for general availability in November 2026.
☁️
Hyperforce on Google Cloud is planned for general availability in North America in November 2026.
📞
Agentforce Voice with Amazon Connect is due later this fall, and OpenAI models through Bedrock are on the way as well.
Also worth knowing
Salesforce kept growing its platform through acquisitions this year, with Contentful joining on September 1 and Fin, formerly known as Intercom, joining on September 10. That is also why one of the new customer agents carries the Fin name.
What actually matters for a mid-sized org
Big keynotes are built for every audience at once. A company with 50 to 500 Salesforce users usually needs only a slice of it, and this table shows where to look first.
If your company is
Focus on first
Sales led with strong inbound demand
Piper and a clean lead routing process
Service heavy with a busy support queue
Casey or Fin together with Multi Agent Orchestration
Running on AWS or Google Cloud
Zero copy data access and wider model choice
Planning 2027 budgets right now
AIforce readiness and a written AI model policy
Three moves for this quarter
🔍
Audit permissions and sharing. AIforce and every new agent run on the access you already have, so old shortcuts become visible much faster.
🎯
Pick one workflow for one agent. Choose a process with a clear owner and a measurable outcome before anyone proposes a whole portfolio.
🧠
Decide who chooses the model. With Koa, Claude, Gemini, OpenAI and Bedrock all on the table, write down which model runs which workload and where your data is allowed to go.
This week we are breaking the biggest topics down in separate articles, starting with AIforce and permissions, then the seven new agents, then Koa and model choice.
Contains information related to marketing campaigns of the user. These are shared with Google AdWords / Google Ads when the Google Ads and Google Analytics accounts are linked together.
90 days
__utma
ID used to identify users and sessions
2 years after last activity
__utmt
Used to monitor number of Google Analytics server requests
10 minutes
__utmb
Used to distinguish new sessions and visits. This cookie is set when the GA.js javascript library is loaded and there is no existing __utmb cookie. The cookie is updated every time data is sent to the Google Analytics server.
30 minutes after last activity
__utmc
Used only with old Urchin versions of Google Analytics and not with GA.js. Was used to distinguish between new sessions and visits at the end of a session.
End of session (browser)
__utmz
Contains information about the traffic source or campaign that directed user to the website. The cookie is set when the GA.js javascript is loaded and updated when data is sent to the Google Anaytics server
6 months after last activity
__utmv
Contains custom information set by the web developer via the _setCustomVar method in Google Analytics. This cookie is updated every time new data is sent to the Google Analytics server.
2 years after last activity
__utmx
Used to determine whether a user is included in an A / B or Multivariate test.
18 months
_ga
ID used to identify users
2 years
_gali
Used by Google Analytics to determine which links on a page are being clicked
30 seconds
_ga_
ID used to identify users
2 years
_gid
ID used to identify users for 24 hours after last activity
24 hours
_gat
Used to monitor number of Google Analytics server requests when using Google Tag Manager