- Agentforce 360 launched: At Dreamforce 2025, Salesforce rolled out Agentforce 360, a new AI-agent platform available globally across its cloud tools [1]. CEO Marc Benioff said Agentforce “connects humans, agents and data on one trusted platform” to automate routine tasks [2].
- ChatGPT & Slack integration: Salesforce is embedding Agentforce into ChatGPT and Slack. An expanded OpenAI partnership lets customers access Agentforce 360 from within ChatGPT (with 800 million weekly users) [3] [4]. At the same time, Slack becomes Salesforce’s new interface – incoming Slack bots and real-time APIs allow ChatGPT and other AI models (like OpenAI’s Codex) to pull Salesforce data and workflows [5] [6].
- Voice and Commerce features: A new Agentforce Voice feature lets AI agents speak naturally over phone systems, integrating with platforms like Amazon Connect and Genesys [7]. Salesforce also announced an “Instant Checkout” commerce agent, built with OpenAI and Stripe’s Agentic Commerce Protocol, to automate online sales and transactions [8].
- Financial results & stock: Salesforce beat Q2 revenue estimates with $10.2 B (+10% YoY) and record margins [9]. CEO Benioff hailed it an “outstanding quarter,” but conservative Q3 guidance caused an immediate sell-off [10]. CRM shares trade in the mid-$240s (around a 52-week low) [11] [12], down ~15–28% for 2025 as investors weigh AI bets and competition. Analysts’ price targets range from ~$221 to $405 [13], reflecting divided outlooks.
- Expert views: Salesforce calls the era “agentic AI,” predicting AI may handle ~40% of Fortune 1000 work alongside humans [14]. But industry research shows ~95% of enterprise AI pilots fail in “pilot purgatory” without proper integration [15]. Experts caution adoption lags expectations: RBC Capital Markets notes “investor enthusiasm around Agentforce has moderated” [16] even as Salesforce doubles down on the platform.
Salesforce’s latest AI push – unveiled at its Oct 2025 Dreamforce conference – centers on Agentforce, a framework of “autonomous AI agents” embedded throughout Salesforce products [17]. With 12,000+ customers already using Agentforce in 2024 [18] [19], the company is betting these agents will break the cycle of stalled pilots. In a press briefing, co-founder CTO Parker Harris said “we are truly in the agentic AI era,” predicting that “40% of the work in the Fortune 1000 is probably going to be done by AI, and it’s going to be humans and AI actually working together” [20]. Salesforce frames Agentforce 360 as the “third wave” of AI – moving beyond chatbots to systems that act on users’ behalf with data and security built-in [21] [22].
New Integrations: ChatGPT, Slack and Voice
A headline announcement is Salesforce’s deepening partnership with OpenAI. Under the expanded alliance, Salesforce customers will soon access Agentforce 360 inside ChatGPT, tapping CRM data and analytics (even building Tableau charts) directly in OpenAI’s interface [23]. Sam Altman, OpenAI’s CEO, called the partnership “an important step in how AI can improve daily workflows” [24]. Practically, that means sales teams could ask ChatGPT for customer history, generate proposals, or complete orders without leaving the chat window. Salesforce also brings ChatGPT into Slack, its workplace messaging tool, to enhance search and insights and even add OpenAI’s Codex code assistant into Slack chats [25].
Slack itself is now being reimagined as the default Salesforce UI [26]. Salesforce is embedding agent apps (for sales, service, HR, etc.) into Slack channels and launching “Channel Expert” – an AI that monitors Slack conversations and answers questions in real time [27]. To enable any AI to use Slack data, Salesforce released a new real-time search API, letting partners (Google, Perplexity, Notion, etc.) build agents inside Slack [28]. As Salesforce engineer Rob Seaman put it, the platform’s power is shown by the AI apps already being built: some startups have signed “tens of thousands of customers” in weeks thanks to Agentforce integration [29].
Salesforce also unveiled Agentforce Voice, letting AI agents speak naturally over phones or voice channels [30]. Built with Amazon Connect, Genesys, and other telephony systems, Agentforce Voice can handle customer calls – updating CRM records, triggering workflows, and handing off to humans when needed [31] [32]. This moves Salesforce beyond text chat into full conversational AI support for contact centers and IT helpdesks. Alongside voice, Salesforce announced an AI-driven commerce agent: using the new “Agentic Commerce Protocol” (developed with Stripe), customers will soon get chatbot help to find products and even a one-click “Instant Checkout” flow [33].
Finally, Salesforce expanded its Anthropic partnership for regulated industries. Claude (Anthropic’s AI) is now an approved agent in Salesforce’s “trust layer,” so finance or healthcare customers can use Claude-powered agents behind Salesforce’s security fences [34]. Claude will appear in Agentforce 360 (especially the new Financial Services edition) and in Slack workspaces for sensitive data analysis [35]. Early case studies include firms like CrowdStrike and RBC Wealth Management running Claude through Salesforce’s Bedrock integration.
Strong Results, Stock Under Pressure
On the financial front, Salesforce still delivered solid growth. In Q2 FY2026 (quarter ended July 2025), revenue was $10.2 billion, up 10% year-over-year – comfortably beating analysts’ estimates [36]. Subscription revenue rose 11%, and operating margins expanded for the 10th consecutive quarter [37]. Marc Benioff praised it as an “outstanding quarter,” noting Salesforce is on track for record operating cash flow [38]. The company returned $2.6 billion to shareholders (dividends and buybacks) in H1 and authorized a huge $20 billion buyback (capitalizing on the depressed share price) [39].
However, guidance tempered enthusiasm. Salesforce projected Q3 (Aug–Oct) revenue of roughly $10.24–10.29B, roughly in line but slightly below consensus [40]. Investors had hoped AI investments would spark faster growth; instead, stock fell about 5% on earnings day and another ~8% after guidance [41]. As of mid-October 2025, CRM shares hover in the mid-$240s. (For context, Franklin Street Advisors noted Salesforce shares around $244.73 – down ~15% over the past year [42] – while Nasdaq is up ~15% in 2025 [43].) Overall, analysts’ price targets now range widely: from bearish mid-$200s to bullish $350–$400 levels [44].
Major firms have mixed views. Goldman Sachs reiterated a Buy on CRM, with a $385 price target, highlighting Data Cloud growth (+120% YoY) and 6,000 paid Agentforce deployments [45]. Morgan Stanley even raised its target to $405 (Overweight) [46]. Conversely, independent analysts and funds are cautious: RBC Capital (via Tekedia) has a Hold rating, noting that “investor enthusiasm around Agentforce has moderated as adoption has lagged expectations” [47]. As one columnist put it, Salesforce’s stock is one of tech’s laggards in 2025 – underperforming AI “winners” like Nvidia and Microsoft [48] – leaving investors to wonder if the AI strategy will soon pay off or continue to face headwinds.
Challenges and the AI Opportunity
Salesforce is not alone in touting AI agents, but it faces key challenges. Industry research shows most AI pilots falter: according to MIT research cited by Salesforce, “95% of enterprise AI pilots fail before production” [49]. The culprit, as Salesforce’s Srini Tallapragada explains, is that disconnected tools and data leave AI models without context, trapping companies in a “prompt doom loop” [50]. Salesforce’s answer is its integrated stack: Data Cloud unifies data, Customer 360 builds in business logic, and Slack ties it together conversationally [51]. In theory, that means an AI agent in Salesforce always knows the latest customer record, past interactions, and company policies as it acts – solving the integration issue.
On the upside, customer case studies are promising. Companies using Agentforce report big productivity gains. For example, Reddit says its AI agents cut average support time from 8.9 to 1.4 minutes (an 84% drop) and deflected 46% of service cases [52]. OpenTable resolved 70% of customer inquiries autonomously with Agentforce, and the travel firm Engine cut support time 15%, saving over $2M annually [53]. Internally, Salesforce’s own helpdesk now handles 1.8 million AI-powered conversations weekly, showing how seriously the company is investing in its platform [54].
Yet investor caution lingers. RBC Capital notes that even Salesforce warned in its earnings that “new AI products may disrupt workforce needs and negatively impact demand” [55]. In other words, companies may hesitate buying more Salesforce seat licenses if AI starts replacing some human work. Analysts also warn that repeated acquisitions (Informatica for $8B, two smaller AI startups, etc. [56]) could invite scrutiny if they don’t quickly boost growth. TS2.tech analyst Rebecca Wettemann (via ts2) cautioned that activist investors might push back if deals fail to accelerate revenue [57].
Outlook and Analysis
Looking ahead, Salesforce believes AI will unlock new markets and efficiency. The company points to a $7 billion “market opportunity” in tasks suitable for automation [58]. CEO Benioff now even describes a vision of a “limitless agentic workforce” where humans and AI agents collaborate seamlessly [59]. To realize that, Salesforce continues rolling out enhancements: Agentforce 3.0 adds more monitoring and industry-specific logic to its agents [60], and Einstein GPT (first launched in 2023) is now integrated into every cloud to auto-generate sales emails, service replies and marketing content [61].
For investors, the coming quarters are critical. Salesforce is hosting an Investor Day on Oct 15, 2025, where management is expected to detail long-term strategy for AI monetization and possibly set new targets [62]. Market watchers will look for evidence that AI-driven products (like Agentforce and Einstein GPT) are translating into renewed growth. As one industry analyst summarized, Salesforce’s future hinges on blending its data and AI stack to finally move the needle beyond past 10% growth: “The real test will be whether AI can drive a new cycle of double-digit revenue growth in coming years.” [63] [64]
In summary, Salesforce has placed a big bet on AI agents – from ChatGPT integrations to voice and commerce – and has solid business momentum. But adoption must catch up with hype. As enterprise AI projects have shown, execution is hard. For now, Salesforce’s stock sits near a multiyear low, reflecting investor skepticism [65] [66]. Whether Agentforce and the “agentic enterprise” vision will vindicate Salesforce’s AI strategy (and reignite stock gains) remains to be seen. What’s clear is that Salesforce is doubling down: it envisions AI agents handling significant parts of business workflows, and it has the cash and partners (OpenAI, Stripe, Anthropic) to go after that vision aggressively [67] [68].
Sources: Authoritative news and analysis from CNBC, Reuters [69], VentureBeat [70], Yahoo Finance (Reuters) [71], and tech industry analysts [72] [73] [74]. Stock data and forecasts from financial research (Motley Fool [75], Capital.com [76]) and expert commentary (RBC via Tekedia [77]).
References
1. www.reuters.com, 2. www.reuters.com, 3. www.constellationr.com, 4. www.constellationr.com, 5. www.constellationr.com, 6. venturebeat.com, 7. www.tekedia.com, 8. www.constellationr.com, 9. ts2.tech, 10. ts2.tech, 11. www.nasdaq.com, 12. www.tekedia.com, 13. capital.com, 14. venturebeat.com, 15. venturebeat.com, 16. www.tekedia.com, 17. venturebeat.com, 18. www.reuters.com, 19. www.tekedia.com, 20. venturebeat.com, 21. www.constellationr.com, 22. venturebeat.com, 23. www.constellationr.com, 24. www.constellationr.com, 25. www.constellationr.com, 26. venturebeat.com, 27. venturebeat.com, 28. venturebeat.com, 29. venturebeat.com, 30. www.tekedia.com, 31. venturebeat.com, 32. www.tekedia.com, 33. www.constellationr.com, 34. www.constellationr.com, 35. www.constellationr.com, 36. ts2.tech, 37. ts2.tech, 38. ts2.tech, 39. ts2.tech, 40. ts2.tech, 41. ts2.tech, 42. www.nasdaq.com, 43. www.tekedia.com, 44. capital.com, 45. capital.com, 46. capital.com, 47. www.tekedia.com, 48. www.nasdaq.com, 49. venturebeat.com, 50. venturebeat.com, 51. venturebeat.com, 52. venturebeat.com, 53. venturebeat.com, 54. venturebeat.com, 55. www.tekedia.com, 56. ts2.tech, 57. ts2.tech, 58. venturebeat.com, 59. ts2.tech, 60. ts2.tech, 61. ts2.tech, 62. ts2.tech, 63. ts2.tech, 64. www.nasdaq.com, 65. www.nasdaq.com, 66. www.tekedia.com, 67. www.constellationr.com, 68. ts2.tech, 69. www.reuters.com, 70. venturebeat.com, 71. www.reuters.com, 72. www.constellationr.com, 73. ts2.tech, 74. www.tekedia.com, 75. www.nasdaq.com, 76. capital.com, 77. www.tekedia.com
