Nicolas Chadeville

Content and Comms @HexagonAB - Fier travailleur de l'Internet, buveur de cafés crèmes par milliers, pire pronostiqueur sportif de l'histoire du sport.

[Nicolas Chadeville @Chad_]

Un cliché américain est que Thanksgiving est l'occasion de grandes disputes familiales entre personnes aux opinions politiques opposées.

Des scientifiques ont testé : en réalité, les débats politiques en famille rapprochent : on en revient plutôt moins polarisé qu'auparavant 👇

[Philipp Heimberger @heimbergecon]

This paper provides survey-based evidence that socialising across political divides can change opinions and reduce political polarisation. Using Thanksgiving as a natural experiment, it shows that people shift toward their families’ views after the holiday.

[Nicolas Chadeville @Chad_]

Pour être en bonne santé, faites 7 000 pas par jour (et non 10 000)

Le mythe de "10 000 pas par jour" naît dans les années 60 au Japon où le chiffre 10 000 (一万) ressemble à un marcheur.

Or la différence se fait en passant de 2 000 et 7 000 pas - impact négligeable au-delà👇

[Nicolas Chadeville @Chad_]

@EarthFirstVoter @daveweigel Conventional wisdom, I believe

[Nicolas Chadeville @Chad_]

@Dr_Draper Is Live And Let Die too obvious?
https://t.co/JMS36Nu3C2

[Nicolas Chadeville @Chad_]

Le management ne fait plus rêver les Français (ni les Allemands, les Américains ou les Japonais)

Part des salariés disant vouloir des responsabilités manageriales :
🇬🇧 65%
🇮🇳 63%
🇪🇸 44%
🇮🇹 39%
🇺🇸 38%
🇫🇷 37%
🇩🇪 31%
🇳🇱 28%
🇯🇵 21%

(Source : Randstad Workmonitor 2024)

[Nicolas Chadeville @Chad_]

Les salariés veulent du cash.

Conclusion de l'étude mondiale d'@ADP - salaire et sécurité sont les désirs n°1 et n°2 des salariés de tout âge et continent.

A rebours des clichés, culture et flexibilité horaire/géo sont loin derrière, avec peu de différence entre générations.

[Nicolas Chadeville @Chad_]

Diagramme du jour :
Ne dites pas (seulement) Net Zero, dites "Nature Positive".

Face à l'extinction des espèces, de plus en plus d'entreprises se fixent l'objectif d'encourager la biodiversité.

Au risque du buzzword : l'engagement est bien plus dur à tracer que les émissions.

[Nicolas Chadeville @Chad_]

Chose extraordinaire que la place des adjectifs : chaque langue a sa règle obscure que peu d'adultes pourraient expliquer.

Mais, à force de lire et d'essayer les combinaisons, chacun comprend "une belle grande voiture rouge", mais tique sur une "grande rouge voiture belle".

[Nicolas Chadeville @Chad_]

Chiffre étonnant du @FT : l'activité "Miles" des compagnies aériennes vaut parfois davantage que la compagnie elle-même.

Un business dans la tourmente : les clients, qui ont engrangé les miles pendant la pandémie, accusent les compagnies de supprimer les avantages promis.

[Nicolas Chadeville @Chad_]

Les Français et BFM TV, je t'aime moi non plus :

La chaîne est la 3ème source d'info des Français en radio/TV + 5ème sur le Web, soit la 1ère marque en cumulé.

C'est aussi la plus décriée : 40% des Français ne lui font pas confiance (vs 32% pour CNews)

via @Reuters Institute

[Nicolas Chadeville @Chad_]

Vos pubs Twitter risquent de n'amener que des robots.

Selon une étude @CHEQ_Inc, les clics des pubs Twitter viendraient à 76% de bots contre 1-2% pour Tiktok ou Instagram.

@mashable évoque aussi des campagnes où Twitter indique des centaines de clics - sans aucun trafic généré.

[Nicolas Chadeville @Chad_]

Pourquoi Amazon et Alexa ont laissé passer le train de l'AI générative 👇

Guerres internes, système où on n'est pas promu en s'attaquant à des problèmes à long terme... Intéressant de voir comment la culture d'entreprise peut amener un acteur dominant à se faire distancer.

[Mihail Eric @mihail_eric]

How Alexa dropped the ball on being the top conversational system on the planet
—
A few weeks ago OpenAI released GPT-4o ushering in a new standard for multimodal, conversational experiences with sophisticated reasoning capabilities.

Several days later, my good friends at PolyAI announced their Series C fundraising round after tremendous growth in the usage of their enterprise voice assistant.

Amid this news, a former Alexa colleague messaged me: You’d think voice assistants would have been our forte at Alexa.

For context, I joined Alexa AI as a research scientist in early 2019. By this time, the Alexa consumer device had existed for 5 years and was already in 100M+ homes throughout the world.

In 2019, Alexa was experiencing a period of hypergrowth. Dozens of new teams sprouted every quarter, huge financial resources were invested, and senior leadership made it clear that Alexa was going to be one of Amazon’s big bets moving forward.

My team was born amidst all this with a simple charter: bring the latest and greatest in AI research into the Alexa product and ecosystem. I’ve often described our group (later dubbed the Conversational Modeling team) as Google Brain meets Alexa AI-SWAT team.

Over the course of the 2.5 years I was there, we grew from 2 to ~20 and tackled every part of the conversational systems stack.

We built the first LLMs for the organization (though back then we didn’t call them LLMs), we built knowledge grounded response generators (though we didn’t call it RAG), and we pioneered prototypes for what it would mean to make Alexa a multimodal agent in your home.

We had all the resources, talent, and momentum to become the unequivocal market leader in conversational AI. But most of that tech never saw the light of day and never received any noteworthy press.

Why?

The reality is Alexa AI was riddled with technical and bureaucratic problems.

Bad Technical Process
–
Alexa put a huge emphasis on protecting customer data with guardrails in place to prevent leakage and access. Definitely a crucial practice, but one consequence was that the internal infrastructure for developers was agonizingly painful to work with.

It would take weeks to get access to any internal data for analysis or experiments. Data was poorly annotated. Documentation was either nonexistent or stale.

Experiments had to be run in resource-limited compute environments. Imagine trying to train a transformer model when all you can get a hold of is CPUs. Unacceptable for a company sitting on one of the largest collections of accelerated hardware in the world.

I remember on one occasion our team did an analysis demonstrating that the annotation scheme for some subset of utterance data was completely wrong, leading to incorrect data labels.

That meant for months our internal annotation team had been mislabeling thousands of data points every single day. When we attempted to get the team to change their annotation taxonomy, we discovered it would require a herculean effort to get even the smallest thing modified.

We had to get the team’s PM onboard, then their manager’s buy-in, then submit a preliminary change request, then get that approved (a multi-month-long process end-to-end).

And most importantly, there was no immediate story for the team’s PM to make a promotion case through fixing this issue other than “it’s scientifically the right thing to do and could lead to better models for some other team.” No incentive meant no action taken.

Since that wasn’t our responsibility and the lift from our side wasn’t worth the effort, we closed that chapter and moved on.

For all I know, they could still be mislabeling those utterances to this day.

Fragmented Org Structures
—
Alexa’s org structure was decentralized by design meaning there were multiple small teams working on sometimes identical problems across geographic locales.

This introduced an almost Darwinian flavor to org dynamics where teams scrambled to get their work done to avoid getting reorged and subsumed into a competing team.

The consequence was an organization plagued by antagonistic mid-managers that had little interest in collaborating for the greater good of Alexa and only wanted to preserve their own fiefdoms.

My group by design was intended to span projects, whereby we found teams that aligned with our research/product interests and urged them to collaborate on ambitious efforts. The resistance and lack of action we encountered was soul-crushing.

I remember on one occasion we were coordinating a project to scale out the large transformers model training I had been leading. This was an ambitious effort which, if done correctly, could have been the genesis of an Amazon ChatGPT (well before ChatGPT was released).

Our Alexa team met with an internal cloud team which independently was initiating similar undertakings. While the goal was to find a way to collaborate on this training infrastructure, over the course of several weeks there were many half-baked promises made which never came to fruition.

At the end of it, our team did our own thing and the sister team did their own thing. Duplicated efforts due to no shared common ground. With no data, infrastructure, or lesson sharing, this inevitably hurt the quality of produced models.

As another example, the Alexa skills ecosystem was Alexa’s attempt to apply Amazonian decentralization to the dialogue problem. Have individual teams own individual skills.

But dialogue is not conducive to that degree of separation of concerns. How can you seamlessly hand off conversational context between skills? This means endowing the system with multi-turn memory (a long-standing dream of dialogue research).

The internal design of the skills ecosystem made achieving this infeasible because each skill acted like its own independent bot. It was conversational AI by an opinionated bot committee each with its own agenda.

Product-Science Misalignment
—
Alexa was viciously customer-focused which I believe is admirable and a principle every company should practice. Within Alexa, this meant that every engineering and science effort had to be aligned to some downstream product.

That did introduce tension for our team because we were supposed to be taking experimental bets for the platform’s future. These bets couldn’t be baked into product without hacks or shortcuts in the typical quarter as was the expectation.

So we had to constantly justify our existence to senior leadership and massage our projects with metrics that could be seen as more customer-facing.

For example, in one of our projects to build an open-domain chat system, the success metric (i.e. a single integer value representing overall conversational quality) imposed by senior leadership had no scientific grounding and was borderline impossible to achieve.

This introduced product/science conflict in every weekly meeting to track the project��s progress leading to manager churn every few months and an eventual sunsetting of the effort.

—
As we look forward, in the battle for the future of the conversational AI market, I still believe it’s anyone’s game.

Today Alexa has sold 500M+ devices, which is a mind-boggling user data moat. But that alone is not enough.

Here’s how I would organize a dialogue systems effort from the ground-up:

Invest in robust developer infrastructure especially around access to compute, data quality assurance, and streamlined data collection processes. Data and compute are the lifeblood of modern ML systems so proactively setting up this foundation is imperative.

Make LLMs the fundamental building block of the dialogue flows. In retrospect, the Alexa skills ecosystem was a premature initiative for the abilities of conversational systems at the time. I liken it to when Leap Motion created and released a developer SDK before the underlying hardware device was stable.

But with the power of modern LLMs, I’m optimistic about redesigning a developer conversational toolkit with LLMs as their primitives.

Ensure product timelines don’t dictate science research time frames. Because things are moving so fast in the AI world, it’s hard not to feel the pressure of shipping quickly. But there are still so many unsolved problems that will take time to solve.

Of course you should conduct research aggressively, but don’t have delivery cycles measured in quarters, as this will produce inferior systems to meet deadlines.

—
If you’re thinking about the future of multimodal conversational systems and interfaces, I would love to hear from you. We’ve got work to do!

[Nicolas Chadeville @Chad_]

Face à la pénurie des vocations, le nucléaire a trouvé une parade temporaire : rappeler ses retraités.

Faut-il s'en réjouir, ou chercher d'autres solutions ?

https://t.co/f8B0tOn09w

[Nicolas Chadeville @Chad_]

Vos destinataires comprennent-ils vos 🙃 ?

Une étude auprès de 2,200 américains répond à une question essentielle : quels sont les emojis les plus incompris ?

Réponse: 💁‍♀️😪💅

[Nicolas Chadeville @Chad_]

Qu'a-t-on perdu en perdant les secrétaires ?

Le numérique a éliminé nombre de postes d'assistants... et déplacé leurs tâches vers les cadres sups eux-mêmes.

Résultat : des managers (cher) payés à des tâches jadis déléguées : slides, emails, organisation des réunions...

[richard shotton @rshotton]

Interesting observation from economist Peter Sassone - time saving software like PowerPoint has potentially reduced company efficiency.

Tasks previously conducted by support staff now done by senior executives.

In @j_amesmarriott column in Times

[Nicolas Chadeville @Chad_]

Attention aux titres racoleurs fondés sur des sondages en ligne.

@pewresearch alerte sur le nombre croissant de 18-29 ans répondant oui à toute question, par trolling ou pour être rémunérés.

Témoin cette question où 12% disent être habilités à piloter un sous-marin nucléaire.

[Nicolas Chadeville @Chad_]

OpenAI dévoile (en accès restreint) SORA, son outil de création de vidéo sur simple instruction texte.

Résultats bluffants, même si l'entreprise reconnaît des bugs (doigts, lettres, mouvements bizarres...) et des inquiétudes à la pelle sur les conséquences d'un tel outil

[Nicolas Chadeville @Chad_]

Au-delà des apparences : belle campagne (de saison) de @depr_hilfe pour sensibiliser au fait que la dépression n'est pas toujours visible.

[Le Monde de L'Énergie @lemondelenergie]

💶La hausse des taux d'intérêt met les énergies renouvelables au défi d’une rentabilité rapide
Une tribune signée Peter Wilson, ingénieur et expert sectoriel de l'énergie au sein d'@HexagonAB
➡️➡️https://t.co/P10Zxr6144

[Nicolas Chadeville @Chad_]

Extraordinaire diagramme : à tous les niveaux de revenus, on est convaincu qu'il suffirait de gagner 30-50% de plus pour être enfin plus heureux et moins stressé.

[Derek Thompson @DKThomp]

Americans at ~every level of income say they need a 30%-50% raise "to feel happy."

Make $50k? "I need about 75k."
Make $75k? "I need about 100k."
Make $150k? "I need 200k."
Make $200k? "I need 350k."