Microsoft drops AI sales targets in half after salespeople miss their quotas

State of AI Hype and Incentives

• Many see the current AI push as driven less by technology readiness and more by greed, FOMO, and financialization: executives chasing “the next big thing” to justify valuations, stock-based comp, and massive capex.
• Others frame it more as herd behavior and fear: no leader wants to be the one who “missed AI” if it does deliver, so they follow industry trends even if they privately doubt the ROI.
• Some argue malice vs incompetence is a false distinction: short‑term profit focus, lack of empathy, and systemic incentives produce the same harmful outcomes either way.

Technical Mismatch and Enterprise Reality

• Commenters repeatedly say current LLMs/“agents” are great at demos and low‑stakes tasks, but not ready for high‑stakes autonomous business workflows.
• The “uncanny valley”: 90–99% correctness is fine in a sales pitch but catastrophic in production (e.g., call centers, legal/financial actions, data operations).
• Hallucinations/confabulation are seen as a fundamental limitation for many enterprise uses; trust and verifiability are bigger blockers than privacy for some.
• Many report that AI often slows them down: they must supervise, correct, and verify, making it easier to “just do it myself.”

Experiences with Microsoft Copilot and Ecosystem

• Strong sentiment that Microsoft’s AI integrations are intrusive and clumsy: constant unwanted autocompletion, bad suggestions in Office, Azure, and IDEs, and frequent need to hit Escape/Undo.
• Copilot and Azure AI are often described as useless or misleading for real troubleshooting and automation; RAG over internal docs is “OK but not good.”
• Some note Microsoft’s long‑standing patterns: bundling weak products, abusing monopoly power, and prioritizing checkbox features over quality, now extended to AI.
• A few counter that Azure OpenAI is seeing significant uptake among larger enterprises because it fits existing contracts and compliance.

Economic and Market Dynamics

• Several see this as early “bubble to trough” behavior: massive infrastructure and GPU spend chasing revenue that isn’t materializing, especially in enterprise.
• Concern that current spending levels are not justified by realistic revenue, and that AI infra may never pay back at current scales or prices.
• Some compare it to past hype cycles (3G, dot‑com, blockchain, EV/“self‑driving”), expecting painful corrections and possible “too big to fail” bailouts.

Cultural, Labor, and Ethical Implications

• Broader critique that tech has shifted culture toward wealth‑maximization and “frictionless” experiences, undermining learning, autonomy, and meaningful effort.
• Fears range from AI as a tool for further domination of capital over labor (mass deskilling, job loss) to more extreme AGI/ASI risk scenarios.
• Others note the danger of “cognitive offloading”: people letting corporations’ models “do the thinking,” similar to social media’s effect on attention and agency.

Disagreement on AI’s Long-Term Importance

• One camp is deeply skeptical: current systems are overhyped, economically marginal, and will mostly produce low‑quality slop plus new failure modes.
• Another camp argues LLMs are genuinely revolutionary (first major change in human–computer interaction since smartphones), with productivity gains already visible in coding, writing, and everyday tasks.
• A middle view: the tech is powerful but current spending is wildly misallocated; most current AI money is being wasted, but significant long‑term value will eventually emerge once real use cases and engineering discipline catch up.