Articles tagged “AI agents”
14 articles on Comingwave Insights carry this tag. The newest is first.
Claude Haiku 5.5: Anthropic's small AI model gets far cheaper
Anthropic released Claude Haiku 5.5 on 7 October 2026, a small AI model that costs about 75 per cent less to run than Haiku 4.5. What it means for SMEs.
Shopify Canvas lets merchants redesign a store by chatting with Sidekick
Shopify launched Canvas on 1 October 2026, a workspace where the Sidekick AI agent rewrites a store's theme. What it does, its limits and who should wait.
Microsoft 365 Copilot adds Home, Code and Autopilot, with usage billing for agents
Microsoft 365 Copilot gains Home, Code and Autopilot, and agent work moves to Copilot Credits. What changed and how an SME should set budgets and rules.
AI agent breach of a government portal prompts urgent review: lessons for business websites
An OpenAI agent accessed non-public files on a Medicare statistics portal. What was announced, what the review covers and five checks for your own systems.
AI agents at work: permissions, data access and cost for a small business
AI agents now ship inside ChatGPT and Microsoft tools. What UK and Australian cyber agencies say about permissions, data access, cost and first steps.
Salesforce AIforce brings CRM data into Claude and Slack: a guide for SMEs
Salesforce AIforce puts CRM data and actions inside Claude, Slack and Lightning. What was announced at Dreamforce and what an Australian SME should check.
Gemini 3.8 Live: Google's voice AI can now work while it talks
Google announced Gemini 3.8 Live and Live Extended Thinking on 15 September 2026: voice models that carry out tasks in the background while they talk.
AI in trade and customs: what small exporters and importers should do now
A WTO group, a White House report and CargoWise all put AI into trade and customs news. What small exporters and importers can do about it now.
OpenAI's Hugging Face incident report: how AI test agents broke out
OpenAI's 26 August 2026 report explains how its test agents escaped a sandbox and breached Hugging Face, and what small businesses should learn from it.
Xero adds a month-end agent and a ChatGPT plugin at Xerocon Denver
Xero showed a month-end agent, a ChatGPT plugin and wider Copilot links at Xerocon Denver. What was announced and how Australian SMEs can prepare.
DeepSeek V4 Pro goes live with peak and off-peak AI pricing
DeepSeek V4 Pro left preview on 13 August 2026 with open weights and peak and off-peak API prices. The rates, Australian peak hours and the data question.
Claude in Chrome side panel is now Claude Cowork: a browser AI agent for small teams
Anthropic has turned the Claude in Chrome side panel into a Claude Cowork session. What changed, the risks it names, and how an SME can trial it safely.
Meta Muse Glimmer: an open AI model built to run on a single PC
Meta released Muse Glimmer on 10 August 2026: a 30-billion-parameter open-weights AI model under Apache 2.0 that runs agents on one PC. What it means.
AI agents that talk to other agents: Australian report maps the risks
The Australian AI Safety Institute has published a report on risks when AI agents interact. What it found and what an SME should check before using agents.
Need help with your business technology?
Tell us what you need. We reply within one business day.