#enterpriseai — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #enterpriseai, aggregated by home.social.
-
Copilot saving time for individuals does not mean Copilot saves time for organisations
I thought this was a really interesting finding from Microsoft’s own research on Copilot 365 using usage data from 6000 trial licenses across 56 firms:
The introduction of Copilot has led to significant changes in email-related behaviors among workers. The average worker in the control group spent 2.8 hours each week reading email and licensees spent 12 fewer minutes reading emails each week, a 7% decrease. This time savings reflects a combination of fewer individual emails read each week, 9 fewer per week, and an increase in reading efficiency, as workers with Copilot spend a few seconds less reading each email on average. Copilot users spent 18% less time reading email, saving more than half an hour each week. We hypothesize that some combination of Outlook Summarize and searching for email content in M365 Copilot Chat allows users to spend less time with individual emails.
If individuals are saving time by reading fewer e-mails and reading existing e-mails more quickly, this plausibly increases the likelihood they will take action which causes more work to be accumulated downstream of the ‘saving’. The authors have an optimistic reading based on Copilot users replying to the same number of e-mails, but the point where ‘more effective triage’ becomes ‘inattention’ is rather ambiguous at best. This doesn’t mean that more time spent reading e-mail is necessarily good, only that less time spent on e-mail is plausibly going to create problems just as much as it represents an efficiency gain. If you just measure efficiency in terms of individualised effects then you’re just not going to pick up on the second-order consequences of AI-generated efficiency savings within organisations. The same point holds true for individuals making documents more quickly:
We expect that the generative capacities of M365 Copilot may help workers produce Word documents more quickly. That speed can have additional ripple effects on how people work. They may have time to contribute to more documents; they may finish rough drafts earlier and take the time to elicit feedback from more colleagues; alternatively, they may co-write less often with colleagues, using Copilot as a substitute writing partner.
This is an interesting finding but there’s absolutely no reason to assume it’s intrinsically a positive thing. People producing documents more quickly can just as easily mean the proliferation of ‘workslop’ in which low quality documents circulate which require more downstream work to address their deficiencies. The social costs only show up in this research design when it comes to meetings:
We looked at Copilot’s effect on the number of Teams meetings attended and total time spent in meetings each week. We find that receiving a Copilot license leads to small and statistically insignificant increases in both metrics, but also a (statistically significant) decrease in the proportion of scheduled meeting time during which people actually met, i.e., Copilot licensees are likelier to join meetings later and leave meetings earlier.
Again not all meetings are valuable. But if Copilot users are likelier to join meetings later and leave earlier it suggest that the AI is being leveraged to support personal time saving, on the assumption the automated features will protect against any determinant to the shared work. It’s not clear to me why this would be the case.
#AI #copilot #efficiency #enterpriseAI #higherEducation #microsoft #university -
Copilot saving time for individuals does not mean Copilot saves time for organisations
I thought this was a really interesting finding from Microsoft’s own research on Copilot 365 using usage data from 6000 trial licenses across 56 firms:
The introduction of Copilot has led to significant changes in email-related behaviors among workers. The average worker in the control group spent 2.8 hours each week reading email and licensees spent 12 fewer minutes reading emails each week, a 7% decrease. This time savings reflects a combination of fewer individual emails read each week, 9 fewer per week, and an increase in reading efficiency, as workers with Copilot spend a few seconds less reading each email on average. Copilot users spent 18% less time reading email, saving more than half an hour each week. We hypothesize that some combination of Outlook Summarize and searching for email content in M365 Copilot Chat allows users to spend less time with individual emails.
If individuals are saving time by reading fewer e-mails and reading existing e-mails more quickly, this plausibly increases the likelihood they will take action which causes more work to be accumulated downstream of the ‘saving’. The authors have an optimistic reading based on Copilot users replying to the same number of e-mails, but the point where ‘more effective triage’ becomes ‘inattention’ is rather ambiguous at best. This doesn’t mean that more time spent reading e-mail is necessarily good, only that less time spent on e-mail is plausibly going to create problems just as much as it represents an efficiency gain. If you just measure efficiency in terms of individualised effects then you’re just not going to pick up on the second-order consequences of AI-generated efficiency savings within organisations. The same point holds true for individuals making documents more quickly:
We expect that the generative capacities of M365 Copilot may help workers produce Word documents more quickly. That speed can have additional ripple effects on how people work. They may have time to contribute to more documents; they may finish rough drafts earlier and take the time to elicit feedback from more colleagues; alternatively, they may co-write less often with colleagues, using Copilot as a substitute writing partner.
This is an interesting finding but there’s absolutely no reason to assume it’s intrinsically a positive thing. People producing documents more quickly can just as easily mean the proliferation of ‘workslop’ in which low quality documents circulate which require more downstream work to address their deficiencies. The social costs only show up in this research design when it comes to meetings:
We looked at Copilot’s effect on the number of Teams meetings attended and total time spent in meetings each week. We find that receiving a Copilot license leads to small and statistically insignificant increases in both metrics, but also a (statistically significant) decrease in the proportion of scheduled meeting time during which people actually met, i.e., Copilot licensees are likelier to join meetings later and leave meetings earlier.
Again not all meetings are valuable. But if Copilot users are likelier to join meetings later and leave earlier it suggest that the AI is being leveraged to support personal time saving, on the assumption the automated features will protect against any determinant to the shared work. It’s not clear to me why this would be the case.
#AI #copilot #efficiency #enterpriseAI #higherEducation #microsoft #university -
Copilot saving time for individuals does not mean Copilot saves time for organisations
I thought this was a really interesting finding from Microsoft’s own research on Copilot 365 using usage data from 6000 trial licenses across 56 firms:
The introduction of Copilot has led to significant changes in email-related behaviors among workers. The average worker in the control group spent 2.8 hours each week reading email and licensees spent 12 fewer minutes reading emails each week, a 7% decrease. This time savings reflects a combination of fewer individual emails read each week, 9 fewer per week, and an increase in reading efficiency, as workers with Copilot spend a few seconds less reading each email on average. Copilot users spent 18% less time reading email, saving more than half an hour each week. We hypothesize that some combination of Outlook Summarize and searching for email content in M365 Copilot Chat allows users to spend less time with individual emails.
If individuals are saving time by reading fewer e-mails and reading existing e-mails more quickly, this plausibly increases the likelihood they will take action which causes more work to be accumulated downstream of the ‘saving’. The authors have an optimistic reading based on Copilot users replying to the same number of e-mails, but the point where ‘more effective triage’ becomes ‘inattention’ is rather ambiguous at best. This doesn’t mean that more time spent reading e-mail is necessarily good, only that less time spent on e-mail is plausibly going to create problems just as much as it represents an efficiency gain. If you just measure efficiency in terms of individualised effects then you’re just not going to pick up on the second-order consequences of AI-generated efficiency savings within organisations. The same point holds true for individuals making documents more quickly:
We expect that the generative capacities of M365 Copilot may help workers produce Word documents more quickly. That speed can have additional ripple effects on how people work. They may have time to contribute to more documents; they may finish rough drafts earlier and take the time to elicit feedback from more colleagues; alternatively, they may co-write less often with colleagues, using Copilot as a substitute writing partner.
This is an interesting finding but there’s absolutely no reason to assume it’s intrinsically a positive thing. People producing documents more quickly can just as easily mean the proliferation of ‘workslop’ in which low quality documents circulate which require more downstream work to address their deficiencies. The social costs only show up in this research design when it comes to meetings:
We looked at Copilot’s effect on the number of Teams meetings attended and total time spent in meetings each week. We find that receiving a Copilot license leads to small and statistically insignificant increases in both metrics, but also a (statistically significant) decrease in the proportion of scheduled meeting time during which people actually met, i.e., Copilot licensees are likelier to join meetings later and leave meetings earlier.
Again not all meetings are valuable. But if Copilot users are likelier to join meetings later and leave earlier it suggest that the AI is being leveraged to support personal time saving, on the assumption the automated features will protect against any determinant to the shared work. It’s not clear to me why this would be the case.
#AI #copilot #efficiency #enterpriseAI #higherEducation #microsoft #university -
Copilot saving time for individuals does not mean Copilot saves time for organisations
I thought this was a really interesting finding from Microsoft’s own research on Copilot 365 using usage data from 6000 trial licenses across 56 firms:
The introduction of Copilot has led to significant changes in email-related behaviors among workers. The average worker in the control group spent 2.8 hours each week reading email and licensees spent 12 fewer minutes reading emails each week, a 7% decrease. This time savings reflects a combination of fewer individual emails read each week, 9 fewer per week, and an increase in reading efficiency, as workers with Copilot spend a few seconds less reading each email on average. Copilot users spent 18% less time reading email, saving more than half an hour each week. We hypothesize that some combination of Outlook Summarize and searching for email content in M365 Copilot Chat allows users to spend less time with individual emails.
If individuals are saving time by reading fewer e-mails and reading existing e-mails more quickly, this plausibly increases the likelihood they will take action which causes more work to be accumulated downstream of the ‘saving’. The authors have an optimistic reading based on Copilot users replying to the same number of e-mails, but the point where ‘more effective triage’ becomes ‘inattention’ is rather ambiguous at best. This doesn’t mean that more time spent reading e-mail is necessarily good, only that less time spent on e-mail is plausibly going to create problems just as much as it represents an efficiency gain. If you just measure efficiency in terms of individualised effects then you’re just not going to pick up on the second-order consequences of AI-generated efficiency savings within organisations. The same point holds true for individuals making documents more quickly:
We expect that the generative capacities of M365 Copilot may help workers produce Word documents more quickly. That speed can have additional ripple effects on how people work. They may have time to contribute to more documents; they may finish rough drafts earlier and take the time to elicit feedback from more colleagues; alternatively, they may co-write less often with colleagues, using Copilot as a substitute writing partner.
This is an interesting finding but there’s absolutely no reason to assume it’s intrinsically a positive thing. People producing documents more quickly can just as easily mean the proliferation of ‘workslop’ in which low quality documents circulate which require more downstream work to address their deficiencies. The social costs only show up in this research design when it comes to meetings:
We looked at Copilot’s effect on the number of Teams meetings attended and total time spent in meetings each week. We find that receiving a Copilot license leads to small and statistically insignificant increases in both metrics, but also a (statistically significant) decrease in the proportion of scheduled meeting time during which people actually met, i.e., Copilot licensees are likelier to join meetings later and leave meetings earlier.
Again not all meetings are valuable. But if Copilot users are likelier to join meetings later and leave earlier it suggest that the AI is being leveraged to support personal time saving, on the assumption the automated features will protect against any determinant to the shared work. It’s not clear to me why this would be the case.
#AI #copilot #efficiency #enterpriseAI #higherEducation #microsoft #university -
Copilot saving time for individuals does not mean Copilot saves time for organisations
I thought this was a really interesting finding from Microsoft’s own research on Copilot 365 using usage data from 6000 trial licenses across 56 firms:
The introduction of Copilot has led to significant changes in email-related behaviors among workers. The average worker in the control group spent 2.8 hours each week reading email and licensees spent 12 fewer minutes reading emails each week, a 7% decrease. This time savings reflects a combination of fewer individual emails read each week, 9 fewer per week, and an increase in reading efficiency, as workers with Copilot spend a few seconds less reading each email on average. Copilot users spent 18% less time reading email, saving more than half an hour each week. We hypothesize that some combination of Outlook Summarize and searching for email content in M365 Copilot Chat allows users to spend less time with individual emails.
If individuals are saving time by reading fewer e-mails and reading existing e-mails more quickly, this plausibly increases the likelihood they will take action which causes more work to be accumulated downstream of the ‘saving’. The authors have an optimistic reading based on Copilot users replying to the same number of e-mails, but the point where ‘more effective triage’ becomes ‘inattention’ is rather ambiguous at best. This doesn’t mean that more time spent reading e-mail is necessarily good, only that less time spent on e-mail is plausibly going to create problems just as much as it represents an efficiency gain. If you just measure efficiency in terms of individualised effects then you’re just not going to pick up on the second-order consequences of AI-generated efficiency savings within organisations. The same point holds true for individuals making documents more quickly:
We expect that the generative capacities of M365 Copilot may help workers produce Word documents more quickly. That speed can have additional ripple effects on how people work. They may have time to contribute to more documents; they may finish rough drafts earlier and take the time to elicit feedback from more colleagues; alternatively, they may co-write less often with colleagues, using Copilot as a substitute writing partner.
This is an interesting finding but there’s absolutely no reason to assume it’s intrinsically a positive thing. People producing documents more quickly can just as easily mean the proliferation of ‘workslop’ in which low quality documents circulate which require more downstream work to address their deficiencies. The social costs only show up in this research design when it comes to meetings:
We looked at Copilot’s effect on the number of Teams meetings attended and total time spent in meetings each week. We find that receiving a Copilot license leads to small and statistically insignificant increases in both metrics, but also a (statistically significant) decrease in the proportion of scheduled meeting time during which people actually met, i.e., Copilot licensees are likelier to join meetings later and leave meetings earlier.
Again not all meetings are valuable. But if Copilot users are likelier to join meetings later and leave earlier it suggest that the AI is being leveraged to support personal time saving, on the assumption the automated features will protect against any determinant to the shared work. It’s not clear to me why this would be the case.
#AI #copilot #efficiency #enterpriseAI #higherEducation #microsoft #university -
AI isn’t magic. In our conversation with Sophie Dionnet (Dataiku), she cuts to the real enterprise AI trap: not models, but change management.
“Taking a decision is one hour, implementation is two years.”
That gap is where AI succeeds or stalls.
Watch on YouTube: https://youtube.com/shorts/Rh2kZ4P_ib4
-
AI isn’t magic. In our conversation with Sophie Dionnet (Dataiku), she cuts to the real enterprise AI trap: not models, but change management.
“Taking a decision is one hour, implementation is two years.”
That gap is where AI succeeds or stalls.
Watch on YouTube: https://youtube.com/shorts/Rh2kZ4P_ib4
-
Anthropic Mulls New AI Model Amid Investors’ Pre-IPO Worries https://www.byteseu.com/2383806/ #AI #AISafety #Anthropic #ArtificialIntelligence #EnterpriseAI #InitialPublicOffering #IPO #News #OpenAI #PYMNTSNews #What'sHot
-
Anthropic Weighs New Model Launch to Counter GPT-6 Astra
Reuters reports Anthropic may release a new model before its IPO, days after CEO Dario Amodei called for the industry to slow down.
-
Infinito.Nexus 14.0: Turn Local AI into a Secure Digital Workforce
In Short Infinito.Nexus 14.0 enables companies to operate AI on their own infrastructure, connect it securely to existing business applications and deploy AI agents such as Hermes and OpenClaw as virtual employees. Sensitive information can be processed by local models, while optional connections to OpenAI, Anthropic and OpenRouter provide access to external frontier models when greater capabilities are needed. Companies decide which data may leave their infrastructure, which tools an agent can use and which actions remain restricted. Infinito.Nexus 14.0 brings models, enterprise applications and autonomous agents together in one controlled infrastructure. Organizations can now combine local AI for sensitive data with external frontier models and deploy agents such as Hermes and OpenClaw as isolated virtual employees. Generative AI has already changed how employees search, write, analyze and develop. But most organizations still face an uncomfortable choice: either send business data to an external AI provider or accept the operational complexity of building a local AI platform from scratch. Infinito.Nexus 14.0 introduces a third option: a hybrid AI architecture in which organizations decide where every workload runs. Sensitive prompts can be processed by local models through Ollama or LM Studio. Tasks that require the capabilities of frontier models can be routed to OpenAI, Anthropic or OpenRouter. Applications and agents use one centrally managed gateway instead of implementing separate provider integrations. The result is not merely another enterprise chatbot. Version 14.0 provides the foundation for a secure digital workforce. […] -
Infinito.Nexus 14.0: Turn Local AI into a Secure Digital Workforce
In Short Infinito.Nexus 14.0 enables companies to operate AI on their own infrastructure, connect it securely to existing business applications and deploy AI agents such as Hermes and OpenClaw as virtual employees. Sensitive information can be processed by local models, while optional connections to OpenAI, Anthropic and OpenRouter provide access to external frontier models when greater capabilities are needed. Companies decide which data may leave their infrastructure, which tools an agent can use and which actions remain restricted. Infinito.Nexus 14.0 brings models, enterprise applications and autonomous agents together in one controlled infrastructure. Organizations can now combine local AI for sensitive data with external frontier models and deploy agents such as Hermes and OpenClaw as isolated virtual employees. Generative AI has already changed how employees search, write, analyze and develop. But most organizations still face an uncomfortable choice: either send business data to an external AI provider or accept the operational complexity of building a local AI platform from scratch. Infinito.Nexus 14.0 introduces a third option: a hybrid AI architecture in which organizations decide where every workload runs. Sensitive prompts can be processed by local models through Ollama or LM Studio. Tasks that require the capabilities of frontier models can be routed to OpenAI, Anthropic or OpenRouter. Applications and agents use one centrally managed gateway instead of implementing separate provider integrations. The result is not merely another enterprise chatbot. Version 14.0 provides the foundation for a secure digital workforce. […] -
Episode Recap: The Three Ingredients That Turn AI Into Value – we ask what enterprise AI success really looks like. Sophie Dionnet says it’s people, orchestration and governance, not just flash tools. She shares how a Roche patent lawyer built his own agents and why controls accelerate scale. Listen on Spotify 👉 https://open.spotify.com/episode/5VHvVxpF3IeM0bNLpuDYFN?si=YPFY_zumTT-Z8h7zO-l8_g&ref=analysepodcast.com #EnterpriseAI #AnalysePodcast
-
Episode Recap: The Three Ingredients That Turn AI Into Value – we ask what enterprise AI success really looks like. Sophie Dionnet says it’s people, orchestration and governance, not just flash tools. She shares how a Roche patent lawyer built his own agents and why controls accelerate scale. Listen on Spotify 👉 https://open.spotify.com/episode/5VHvVxpF3IeM0bNLpuDYFN?si=YPFY_zumTT-Z8h7zO-l8_g&ref=analysepodcast.com #EnterpriseAI #AnalysePodcast