Mediaboss Marketing

TAMPA, Fla. Amazon Web Services is busy positioning its cloud infrastructure business to capitalize on the promise of generative artificial intelligence for transforming space and other industries.

More than 60% of the companys space and aerospace customers are already using some form of AI in their businesses, according to AWS director of aerospace and satellite Clint Crosier, up from single digits around three years ago.

Crosier predicts similar growth over the next few years in space for generative AI, which uses deep-learning models to answer questions or create content based on patterns detected in massive datasets, marking a major step up from traditional machine-learning algorithms.

Mathematical advances, an explosion in the amount of available data and cheaper and more efficient chips for processing it are a perfect storm for the rise of generative AI, he told SpaceNews in an interview, helping drive greater adoption of cloud-based applications.

In the last year, AWS has fundamentally reorganized itself internally so that we could put the right teams [and] organizational structure in place so that we can really double down on generative AI, he said.

He said AWS has created a generative AI for space cell of a handful of people to engage with cloud customers to help develop next-generation capabilities.

These efforts include a generative AI laboratory for customers to experiment with new ways of using these emerging capabilities.

Crosier sees three main areas for using generative AI in space: geospatial analytics, spacecraft design and constellation management.

Earth observation satellite operators such as BlackSky and Capella Space already use these tools to help manage search queries and gain more insights into their geospatial data.

Its early days in the manufacturing sector, but Crosier said engineers are experimenting with how a generative AI model fed with design parameters could produce new concepts by drawing from potentially overlooked data, such as from the automotive industry.

Whether youre designing a satellite, rocket or spacecraft, youre letting the generative AI go out and do that exploratory work around the globe with decades of data, he said, and then it will come back and bring you novel design concepts that nobody has envisioned before for your team to use as a baseline to start refining.

He said generative AI also has the potential to help operators manage increasingly crowded orbits by helping to simulate testing scenarios.

If I have a constellation of 600 satellites, I want to model how that constellation will behave under various design parameters, he said.

Well, I can get a model of two concepts, which leaves me woefully inadequate but it costs time and money to model them, or I can model an infinite number. Gen AI will tell me what are the top 25 cases I should model for my modeling simulation capability that will give me the best design optimization, and so were seeing it used that way.

AWS efforts to accelerate the adoption of these emerging computing capabilities also include scholarships and a commitment announced in November to provide free AI training for two million people worldwide before the end of 2025.

Continue reading here:
Bringing generative artificial intelligence to space - SpaceNews

In the latest episode of Lexicon, the podcast by Interesting Engineering (IE), we sit down with Jaroslaw Rzepecki, Ph. D., Monumos chief technology officer (CTO).

Our mission is to improve the efficiency of motor systems in a way that has never been possible before and, in doing so, help us use precious resources more sustainably, Monumo explains. Monumo is working hard to get there using a unique set of data and machine learning techniques to build one of the worlds first large engineering models (LEM).

Once matured, this model will work like an engineering R&D Midjourney or Dall-E to help engineers with components or entire machine plans on demand. A quantum leap in computer-aided design (CAD), if you like.

While the interface wont be as dumbed down as you might expect from large-language models (LLMs) like ChatGPT, it will leverage an engineers time to make the best kit they can imagine. And the potential is enormous.

Jaroslaw Rzepecki leads the companys technological development, oversees the hardware and software development pipelines, and directs machine learning (ML) research.

Before joining Monumo, Jaroslaw was an integral part of the Codemasters team behind the racing video games Grid and Dirt 2; he has also worked as a software engineer at Siemens and held senior roles at Microsoft Research and ARM.

As he told Interesting Engineering during our interview, he also spends some of his spare time in martial arts, specifically kickboxing. We asked him if martial arts had helped his professional life.

Yeah, so its a bit similar to my professional journey, so I tried several different disciplines as I moved around. You know I was also changing clubs, and obviously, then you also change the styles a little bit, Jaroslaw told us.

I did quite a few different ones. I would say that my favorite sport is kickboxing. Ive done that for probably the longest out of all of them, and whether it helps, it does. I think it helps with focus. It helps with clearing your mind, he added.

Afterwards, you probably feel physically exhausted; youre quite invigorated. You have more energy that day to do something than if you would skip that training the previous day. So yes, I would say that it does help, Jaroslaw said.

After his extensive and diverse career, including academia, computer game design, and software engineering at Siemens and ARM, Jaroslaw saw the potential for Monumo and jumped ship to become its second-ever employee. He has since worked up the ranks to become its head tech honcho.

When asked if this was a big risk for him, Jaroslaw said, Um, there is always some risk involved when you change, right? But you know no risk, no fun, right? So, yes, I think a bit of a risk was involved. But um, as I said, I calculated that risk and thought, thats okay.

The main thrust of Monumos work is to combine physics and engineering knowledge with machine learning (ML) and artificial intelligence (AI) to build a computer model that can help sketch out new models for machines. The idea is that, with enough data and training, such a model could conceivably be used to make novel designs never dreamed up before.

And it will be data and professional-driven to boot. Not just any Tom, Dick, and Harry will be able to pick it up and run with it. This is mainly because Monumo plans to keep its software proprietary but also because, at its heart, the software is a complicated multidisciplinary physics model.

It combines data and understanding of many different engineering fields and physics and can conceivably integrate many other diverse fields. This could encompass nuclear physics, nanotechnology, biology, and geology. It could be integrated into the model if it can be measured or modeled.

One sentence headline here, and Im sure that everybody in the engineering community listening to this podcast will appreciate how difficult it is to find the right balance of different components of a complex engineering system if you want to design it, right? Jaroslaw said.

Its a difficult problem. So I like a challenge, I like difficult problems, and applying deep tech to engineering also automatically makes it a multidisciplinary problem because obviously, you have to combine, you know, the latest developments in computer science algorithms optimization, math, and physics, he added.

But the LEM is the long-term goal. For now, they are building an Anser model that can generate models but, crucially, provide the training data for the LEM later down the line. Monumo is focusing on making electric motors as energy-efficient as possible.

When pressed about problems of LLMs and hallucinations, Jaroslaw explained that Anser and the eventual LEM would be immune to this. He explained this because the generated designs are then sense checked using mechanical engineering tools to assess their viability.

If they dont pass the muster, the software flags issues, and the user will go back to the drawing board to amend the design accordingly. The entire design process is the same as in real life, with multiple stages yielding the final piece.

It is a collaborative approach, like tweaking parameters in Midjourney or Dall-E to get the picture you want. Anser can also integrate certain customer considerations or constraints into the design based on their needs.

Since many aspects of our modern world use energy in some form or another, even a marginal increase in energy efficiency could provide enormous energy savings around the world. Less energy wasted is a bonus for the planet as a whole.

And so any kind of improvements that we can make to electric motors will have a huge positive impact on ecology and our movement of the society to towards a more and more green way of life, Jaroslaw said.

The company chose the electric motor as it is a simple and complex enough problem. If Anser can be proven with something like this, it can be used for basically anything (within reason) with enough data and training.

The techniques that were applying and the simulation that we build is a multiphysics simulation so that it could be applied to other branches of engineering we are indeed. Yes, we are laying the foundations and building the simulation that is flexible enough, he explained.

LLMs (Large Language Models) drive todays AI models to mimic human ability with words and pictures. Tomorrow, LEMs (Large Engineering Models) will create solutions that surpass anything humans have previously achieved. Our ability to run and store large volumes of simulations, combined with our optimization intelligence, means that we are already on the way to building these precious data sets and training new models, Monumo explains.

And dont worry about such a model taking your engineering job. Jaroslaw explained that Anser and its progeny should be considered a new, competent computer-aided (CAD) design software.

I dont think that Engineers have to worry about losing their jobs. I will always need engineers. You know, all of this is, um, its a tool, and weve seen in the past that each time a new tool is developed in principle, he said.

Humankind has an option: either Im going to use this great new tool and do the same thing that I did before but with fewer humans being involved, or I can use this new tool and all the humans that I have just to do more, and we always go for Lets just do more, he added.

So, it may be time to brush up on your AI and ML expertise.

NEWSLETTER

Stay up-to-date on engineering, tech, space, and science news with The Blueprint.

Christopher McFadden Christopher graduated from Cardiff University in 2004 with a Masters Degree in Geology. Since then, he has worked exclusively within the Built Environment, Occupational Health and Safety and Environmental Consultancy industries. He is a qualified and accredited Energy Consultant, Green Deal Assessor and Practitioner member of IEMA. Chris’s main interests range from Science and Engineering, Military and Ancient History to Politics and Philosophy.

Read more:
Here's how AI and ML are shaping the future of machine design - Interesting Engineering

One of the hottest investment trends to jump on at the moment is machine learning stocks to buy. Valued at about $79.3 billion at the moment, its expected to balloon to$503.4 billion by the time 2030rolls around, according to Statista.

All thanks to demand for accurate prediction and better decision making for companies and governments of all sizes. Were also seeing machine learning companies pop up in healthcare, finance, security and retail to name a few industries.

Along the way, machines will learn from historical data, identify patterns, and make logical decisions with little to no need for human interaction. Look at healthcare, for example. Its helping with faster data collection through wearables that machines can learn from. Its helping with accelerated drug discovery and development.

Plus,as noted by BuiltIn.com, By crunching large volumes of data,machine learning technology can help healthcare professionalsgenerate precise medicine solutions customized to individual characteristics. Machine learning models can also predict how patients react to certain drugs, allowing healthcare workers to proactively address patients needs.

We could easily go on. But you can see why were excited about machine learning, and the significant impact it will have on just about everything.So, how can we profit from it all? Here are three machine learning stocks you may want to buy.

Source: Sisacorn / Shutterstock.com

The last time I mentionedLantern Pharma(NASDAQ:LTRN), it traded at $5.25on May 1.

At the time, I noted, An artificial intelligence company, its helping to transform the cost and speed to oncology drug discovery and development with itsAI and machine learning platform, RADR.With the help of machine learning, AI and advanced genomics, its platform can scan billions of data points to help identity compounds that could help cancer patients.

Now trading at $6.46, theres even more to get excited about.

For one,Lantern just received regulatory approval to expand its Harmonic trial,a Phase 2 trial thats evaluating LP-300 fornon-small cell lung cancer, or NSCLC, in people that have never smoked in Japan and Taiwan. About athird of all lung cancer patientsin East Asia have never smoked, with numbers still rising.

With the expanded study, it can accelerate the collection of patient and response data needed for the development of LP-300. And if successful, the treatment may be able to help treat relapsed and inoperable adenocarcinoma of the lung in combination with chemotherapy.

Its also working with French biotech company,Oregon Therapeuticsto developprotein disulfide isomerase, or PDI, inhibitor drug candidate XCE853. Lantern will use its RADR AI platform to uncover biomarkers and efficacy-associated signatures of XCE853 across solid tumors that can aid in precision development,as noted in a company press release.

Source: Gorodenkoff / Shutterstock.com

We can also look atExscientia(NASDAQ:EXAI), an AI-driven precision medical company thats accelerating drug development and reducing the time to get it to market.

In fact,as noted by the company, Exscientia developed the first-ever functional precision oncology platform to successfully guide treatment selection and improve patient outcomes in a prospective interventional clinical study, as well as to progress AI-designed small molecules into the clinical setting.

At the moment, the company in still in Phase 1/2 studies for GTAEXS617, a potential best in class CDK7 inhibitor for the treatment of solid tumors. The company expects to move into a combination phase of the trial by the second half of the year.

In addition,EXS74539 is the companys LSD1 inhibitorand is currently making its way through IND-CTA-enabling studies (investigational new drug-clinical trial application). With it, EXAI plans to submit an IND or CTA by the third quarter of the year. It also has a goal to initiate a Phase 1/2 trial for acute myeloid leukemia (AML) patients by the end of the year.

Source: Shutterstock

Or, if you want to diversify with AI-focused companies that will benefit from AI and machine learning, theres theRoundhill Generative AI & Technology ETF(NYSEARCA:CHAT).

With an expense ratio of 0.75%, the ETF holds 50 related stocks, such asNvidia(NASDAQ:NVDA),Microsoft(NASDAQ:MSFT),Alphabet(NASDAQ:GOOG),Meta Platforms(NASDAQ:META),Advanced Micro Devices(NASDAQ:AMD), andAdobe(NASDAQ:ADBE) to name a few. All of which stand to benefit from the artificial intelligence and machine learning story.

Even better, I can buy 100 shares of CHAT for about $3,500, and gain exposure to those 50 holdings. Thats far better than buying just one of its holdings lets say 100 shares of just NVDA for about $95,000.

With the ETF, youre diversified and all your eggs arent in just one basket.

On the date of publication, Ian Cooper did not hold (either directly or indirectly) any positions in the securities mentioned. The opinions expressed in this article are those of the writer, subject to the InvestorPlace.comPublishing Guidelines.

Ian Cooper, a contributor to InvestorPlace.com, has been analyzing stocks and options for web-based advisories since 1999.

See more here:
The 3 Best Machine Learning Stocks to Quadruple Your Money by 2035 - InvestorPlace

Krakauer, J. W., Ghazanfar, A. A., Gomez-Marin, A., MacIver, M. A. & Poeppel, D. Neuroscience needs behavior: correcting a reductionist bias. Neuron 93, 480490 (2017).

Article CAS PubMed Google Scholar

Anderson, D. J. & Perona, P. Toward a science of computational ethology. Neuron 84, 1831 (2014).

Article CAS PubMed Google Scholar

Egnor, S. E. R. & Branson, K. Computational analysis of behavior. Annu. Rev. Neurosci. 39, 217236 (2016).

Article CAS PubMed Google Scholar

Datta, S. R., Anderson, D. J., Branson, K., Perona, P. & Leifer, A. Computational neuroethology: a call to action. Neuron 104, 1124 (2019).

Article CAS PubMed PubMed Central Google Scholar

Falkner, A. L., Grosenick, L., Davidson, T. J., Deisseroth, K. & Lin, D. Hypothalamic control of male aggression-seeking behavior. Nat. Neurosci. 19, 596604 (2016).

Article CAS PubMed PubMed Central Google Scholar

Ferenczi, E. A. et al. Prefrontal cortical regulation of brainwide circuit dynamics and reward-related behavior. Science 351, aac9698 (2016).

Article PubMed PubMed Central Google Scholar

Kim, Y. et al. Mapping social behavior-induced brain activation at cellular resolution in the mouse. Cell Rep. 10, 292305 (2015).

Article CAS PubMed Google Scholar

Gunaydin, L. A. et al. Natural neural projection dynamics underlying social behavior. Cell 157, 15351551 (2014).

Article CAS PubMed PubMed Central Google Scholar

Graving, J. M. et al. DeepPoseKit, a software toolkit for fast and robust animal pose estimation using deep learning. eLife 8, e47994 (2019).

Article CAS PubMed PubMed Central Google Scholar

Mathis, A. et al. DeepLabCut: markerless pose estimation of user-defined body parts with deep learning. Nat. Neurosci. 21, 12811289 (2018).

Article CAS PubMed Google Scholar

Pereira, T. D. et al. Fast animal pose estimation using deep neural networks. Nat. Methods 16, 117125 (2019).

Article CAS PubMed Google Scholar

Geuther, B. Q. et al. Robust mouse tracking in complex environments using neural networks. Commun. Biol. 2, 124 (2019).

Article PubMed PubMed Central Google Scholar

Gris, K. V., Coutu, J.-P. & Gris, D. Supervised and unsupervised learning technology in the study of rodent behavior. Front. Behav. Neurosci. 11, 141 (2017).

Schaefer, A. T. & Claridge-Chang, A. The surveillance state of behavioral automation. Curr. Opin. Neurobiol. 22, 170176 (2012).

Article CAS PubMed PubMed Central Google Scholar

Robie, A. A., Seagraves, K. M., Egnor, S. E. R. & Branson, K. Machine vision methods for analyzing social interactions. J. Exp. Biol. 220, 2534 (2017).

Article PubMed Google Scholar

Vu, M.-A. T. et al. A shared vision for machine learning in neuroscience. J. Neurosci. 38, 16011607 (2018).

Article CAS PubMed PubMed Central Google Scholar

Goodwin, N. L., Nilsson, S. R. O., Choong, J. J. & Golden, S. A. Toward the explainability, transparency, and universality of machine learning for behavioral classification in neuroscience. Curr. Opin. Neurobiol. 73, 102544 (2022).

Article CAS PubMed PubMed Central Google Scholar

Newton, K. C. et al. Lateral line ablation by ototoxic compounds results in distinct rheotaxis profiles in larval zebrafish. Commun. Biol. 6, 115 (2023).

Article Google Scholar

Jernigan, C. M., Stafstrom, J. A., Zaba, N. C., Vogt, C. C. & Sheehan, M. J. Color is necessary for face discrimination in the Northern paper wasp, Polistes fuscatus. Anim. Cogn. 26, 589598 (2022).

Article PubMed PubMed Central Google Scholar

Dahake, A. et al. Floral humidity as a signal not a cue in a nocturnal pollination system. Preprint at bioRxiv https://doi.org/10.1101/2022.04.27.489805 (2022).

Dawson, M. et al. Hypocretin/orexin neurons encode social discrimination and exhibit a sex-dependent necessity for social interaction. Cell Rep. 42, 112815 (2023).

Article CAS PubMed Google Scholar

Baleisyte, A., Schneggenburger, R. & Kochubey, O. Stimulation of medial amygdala GABA neurons with kinetically different channelrhodopsins yields opposite behavioral outcomes. Cell Rep. 39, 110850 (2022).

Article CAS PubMed Google Scholar

Cruz-Pereira, J. S. et al. Prebiotic supplementation modulates selective effects of stress on behavior and brain metabolome in aged mice. Neurobiol. Stress 21, 100501 (2022).

Article CAS PubMed PubMed Central Google Scholar

Linders, L. E. et al. Stress-driven potentiation of lateral hypothalamic synapses onto ventral tegmental area dopamine neurons causes increased consumption of palatable food. Nat. Commun. 13, 6898 (2022).

Article CAS PubMed PubMed Central Google Scholar

Slivicki, R. A. et al. Oral oxycodone self-administration leads to features of opioid misuse in male and female mice. Addiction Biol. 28, e13253 (2023).

Article CAS Google Scholar

Miczek, K. A. et al. Excessive alcohol consumption after exposure to two types of chronic social stress: intermittent episodes vs. continuous exposure in C57BL/6J mice with a history of drinking. Psychopharmacology (Berl.) 239, 32873296 (2022).

Article CAS PubMed Google Scholar

Cui, Q. et al. Striatal direct pathway targets Npas1+ pallidal neurons. J. Neurosci. 41, 39663987 (2021).

Article CAS PubMed PubMed Central Google Scholar

Chen, J. et al. A MYT1L syndrome mouse model recapitulates patient phenotypes and reveals altered brain development due to disrupted neuronal maturation. Neuron 109, 37753792 (2021).

Article CAS PubMed PubMed Central Google Scholar

Rigney, N., Zbib, A., de Vries, G. J. & Petrulis, A. Knockdown of sexually differentiated vasopressin expression in the bed nucleus of the stria terminalis reduces social and sexual behaviour in male, but not female, mice. J. Neuroendocrinol. 34, e13083 (2021).

Winters, C. et al. Automated procedure to assess pup retrieval in laboratory mice. Sci. Rep. 12, 1663 (2022).

Article CAS PubMed PubMed Central Google Scholar

Neira, S. et al. Chronic alcohol consumption alters home-cage behaviors and responses to ethologically relevant predator tasks in mice. Alcohol Clin. Exp. Res. 46, 16161629 (2022).

Article PubMed PubMed Central Google Scholar

Kwiatkowski, C. C. et al. Quantitative standardization of resident mouse behavior for studies of aggression and social defeat. Neuropsychopharmacology 46, 15841593 (2021).

Yamaguchi, T. et al. Posterior amygdala regulates sexual and aggressive behaviors in male mice. Nat. Neurosci. 23, 11111124 (2020).

Article CAS PubMed PubMed Central Google Scholar

Nygaard, K. R. et al. Extensive characterization of a Williams syndrome murine model shows Gtf2ird1-mediated rescue of select sensorimotor tasks, but no effect on enhanced social behavior. Genes Brain Behav. 22, e12853 (2023).

Article CAS PubMed PubMed Central Google Scholar

Ojanen, S. et al. Interneuronal GluK1 kainate receptors control maturation of GABAergic transmission and network synchrony in the hippocampus. Mol. Brain 16, 43 (2023).

Article CAS PubMed PubMed Central Google Scholar

Hon, O. J. et al. Serotonin modulates an inhibitory input to the central amygdala from the ventral periaqueductal gray. Neuropsychopharmacology 47, 21942204 (2022).

Article CAS PubMed PubMed Central Google Scholar

Murphy, C. A. et al. Modeling features of addiction with an oral oxycodone self-administration paradigm. Preprint at bioRxiv https://doi.org/10.1101/2021.02.08.430180 (2021).

Neira, S. et al. Impact and role of hypothalamic corticotropin releasing hormone neurons in withdrawal from chronic alcohol consumption in female and male mice. J. Neurosci. 43, 76577667 (2023).

Article CAS PubMed PubMed Central Google Scholar

Lapp, H. E., Salazar, M. G. & Champagne, F. A. Automated maternal behavior during early life in rodents (AMBER) pipeline. Sci. Rep. 13, 18277 (2023).

Article CAS PubMed PubMed Central Google Scholar

Barnard, I. L. et al. High-THC cannabis smoke impairs incidental memory capacity in spontaneous tests of novelty preference for objects and odors in male rats. eNeuro 10, ENEURO.0115-23.2023 (2023).

Article PubMed PubMed Central Google Scholar

Ausra, J. et al. Wireless battery free fully implantable multimodal recording and neuromodulation tools for songbirds. Nat. Commun. 12, 1968 (2021).

Article CAS PubMed PubMed Central Google Scholar

Friard, O. & Gamba, M. BORIS: a free, versatile open-source event-logging software for video/audio coding and live observations. Methods Ecol. Evol. 7, 13251330 (2016).

Article Google Scholar

Spink, A. J., Tegelenbosch, R. A. J., Buma, M. O. S. & Noldus, L. P. J. J. The EthoVision video tracking systema tool for behavioral phenotyping of transgenic mice. Physiol. Behav. 73, 731744 (2001).

Article CAS PubMed Google Scholar

Lundberg, S. shap. https://github.com/shap/shap

Lauer, J. et al. Multi-animal pose estimation, identification and tracking with DeepLabCut. Nat. Methods 19, 496504 (2022).

Article CAS PubMed PubMed Central Google Scholar

Pereira, T. D. et al. SLEAP: a deep learning system for multi-animal pose tracking. Nat Methods 19, 486495 (2022).

Segalin, C. et al. The Mouse Action Recognition System (MARS) software pipeline for automated analysis of social behaviors in mice. eLife 10, e63720 (2021).

Article CAS PubMed PubMed Central Google Scholar

Breiman, L. Random forests. Mach. Learn. 45, 532 (2001).

Article Google Scholar

Liaw, A. & Wiener, M. Classification and regression by randomForest. R News 2/3 https://journal.r-project.org/articles/RN-2002-022/RN-2002-022.pdf (2022).

Goodwin, N. L., Nilsson, S. R. O. & Golden, S. A. Rage against the machine: advancing the study of aggression ethology via machine learning. Psychopharmacology 237, 25692588 (2020).

Article CAS PubMed PubMed Central Google Scholar

Lundberg, S. M. et al. From local explanations to global understanding with explainable AI for trees. Nat. Mach. Intell. 2, 5667 (2020).

Article PubMed PubMed Central Google Scholar

Ribeiro, M. T., Singh, S., & Guestrin, C. Why should I trust you?: explaining the predictions of any classifier. Preprint at arXiv https://doi.org/10.48550/arXiv.1602.04938 (2016).

Sundararajan, M., Taly, A. & Yan, Q. Axiomatic attribution for deep networks. In Proc. of the 34th International Conference on Machine Learning 33193328 (MLR Press, 2017).

Hatwell, J., Gaber, M. M. & Azad, R. M. A. CHIRPS: explaining random forest classification. Artif. Intell. Rev. 53, 57475788 (2020).

Article Google Scholar

Lundberg, S. & Lee, S.-I. A unified approach to interpreting model predictions. Preprint at arXiv https://doi.org/10.48550/arXiv.1705.07874 (2017).

Verma, S., Dickerson, J. & Hines, K. Counterfactual explanations for machine learning: a review. Preprint at arXiv https://doi.org/10.48550/arXiv.2010.10596 (2020).

Read the original post:
Simple Behavioral Analysis (SimBA) as a platform for explainable machine learning in behavioral neuroscience - Nature.com

Slack has been using customer data to power its machine learning functions, including search result relevance and ranking, leading to the company being criticized over confusing policy updates that led many to believe that their data was being used to train its AI models.

According to the company's policy, those wishing to opt out must do so through their organizations Slack admin, who must email the company to put a stop to data use.

Slack has confirmed in correspondence to TechRadar Pro that the information it uses to power its ML not its AI is de-identified and does not access message content.

An extract from the companys privacy principles page reads:

To develop non-generative AI/ML models for features such as emoji and channel recommendations, our systems analyze Customer Data (e.g. messages, content, and files) submitted to Slack as well as Other Information (including usage information) as defined in our Privacy Policy and in your customer agreement.

Another passage reads: To opt out, please have your org, workspace owners or primary owner contact our Customer Experience team at feedback@slack.com

The company does not provide a timeframe for processing such requests.

Sign up to the TechRadar Pro newsletter to get all the top news, opinion, features and guidance your business needs to succeed!

In response to uproar among the community, the company posted a separate blog post to address concerns arising, adding: We do not build or train these models in such a way that they could learn, memorize, or be able to reproduce any customer data of any kind.

Slack confirmed that user data is not shared with third-party LLM providers for training purposes.

The company added in its correspondence to TechRadar Pro that its "intelligent features (not Slack AI) analyze metadata like user behavior data surrounding messages, content and files but they don't access message content."

Here is the original post:
Slack is training its machine learning on your chat behavior unless you opt out via email - TechRadar